Background

Acetaminophen and 5-hydroxytryptamine type 3 (5-HT3) receptor antagonists are administered as standard prophylaxes for postoperative pain, nausea, and vomiting. Preclinical studies, however, suggest that 5-HT3 antagonists may compromise acetaminophen’s analgesic effect. This hospital registry study investigates whether 5-HT3 antagonists mitigate the analgesic effect of prophylactic acetaminophen in a perioperative setting.

Methods

This study included 55,016 adult patients undergoing general anesthesia for ambulatory procedures at a tertiary healthcare center in Massachusetts from 2015 to 2022. Using binary exposure variables and a comprehensive selection of preplanned patient- and procedure-related covariates for confounder control, the authors investigated whether intraoperative 5-HT3 antagonists affected the association between pre- or intraoperative acetaminophen and postoperative opioid consumption, gauged by opioid dose in milligram oral morphine equivalents (OME) administered in the postanesthesia care unit. A multivariable, zero-inflated negative binomial regression model was applied.

Results

A total of 3,166 patients (5.8%) received only acetaminophen, 15,438 (28.1%) only 5-HT3 antagonists, 31,850 (57.9%) both drugs, and 4,562 (8.3%) neither drug. The median postanesthesia care unit opioid dose was 7.5 mg OME (interquartile range, 7.5 to 14.3 mg OME) among 16,640 of 55,016 (30.2%) patients who received opioids, and the mean opioid dose was 3.2 mg OME across all patients (maximum cumulative dose, 20.4 mg OME). Acetaminophen administration was associated with a –5.5% (95% CI, –9.6 to –1.4%; P = 0.009; adjusted absolute difference, –0.19 mg OME; 95% CI, –0.33 to –0.05; P = 0.009) reduction in opioid consumption among patients who did not receive a 5-HT3 antagonist, while there was no effect in patients who received a 5-HT3 antagonist (adjusted absolute difference, 0.00 mg OME; 95% CI, –0.06 to 0.05; P = 0.93; P for interaction = 0.013).

Conclusions

A dose-dependent association of pre- or intraoperative acetaminophen with decreased postoperative opioid consumption was not observed when 5-HT3 antagonists were coadministered, suggesting that physicians might consider reserving 5-HT3 antagonists as rescue medication for postoperative nausea or vomiting when acetaminophen is administered for pain prophylaxis.

Editor’s Perspective
What We Already Know about This Topic
  • Acetaminophen and 5-hydroxytryptamine type 3 (5-HT3) antagonists are some of the most widely used perioperative medications

  • One proposed analgesic mechanism of action for acetaminophen is 5-HT3 receptor agonism, with previous preclinical and healthy volunteer studies showing decreased analgesia in the presence of 5-HT3 antagonists

What This Article Tells Us That Is New
  • This hospital registry study of 55,016 ambulatory surgery patients found that a dose-dependent reduction in postanesthesia care unit opioid consumption associated with pre- or intraoperative acetaminophen was not observed in cases in which a 5-HT3 antagonist was coadministered

  • This apparent modification of the analgesic effect of acetaminophen was not observed with other commonly used antiemetics, such as dexamethasone or haloperidol

  • Although this apparent reduction of analgesic effect persisted when a host of prespecified covariates were controlled for, a definitive randomized controlled trial is needed to confirm this effect and delineate any clinical relevance to surgical and patient subgroups

Acetaminophen is one of the most commonly used nonopioid analgesic drugs worldwide with more than 60 million weekly users in the United States and an estimated global market value of almost $10 billion in 2022.1,2  In patients undergoing surgery, acetaminophen is frequently used in multimodal, opioid-sparing anesthetic regimens for prophylaxis of postoperative pain.3  The drug’s widespread use and multitude of potential mechanisms of action underline the need to investigate potential interactions between acetaminophen and other drugs.4  Previous studies suggest that the analgesic effects of acetaminophen involve activation of the spinal descending serotonergic system.5–7  This raises the question of whether acetaminophen interacts with medications modulating serotonergic systems, including 5-hydroxytryptamine type 3 (5-HT3) receptor antagonists used for prophylaxis and treatment of postoperative nausea and vomiting. Preclinical animal studies and investigations in healthy volunteers reported that 5-HT3 receptor antagonists mitigate the analgesic effect of acetaminophen.5–8  While these observations were interpreted as proof of the potential mechanism of action of acetaminophen, little consideration was given to their clinical implications. In the perioperative setting, acetaminophen as prophylaxis for postoperative pain and 5-HT3 antagonists as prophylaxis for postoperative nausea and vomiting are combined in a large proportion of patients. With approximately 300 million patients worldwide undergoing general anesthesia annualy,9  a clinically relevant interaction may thus impact a vast population of patients. However, whether the interaction between acetaminophen and 5-HT3 antagonists is relevant in a clinical setting remains uncertain.

In this hospital registry study, we hypothesized that the coadministration of 5-HT3 antagonists for prophylaxis of postoperative nausea and vomiting mitigates the analgesic effect of acetaminophen and increases opioid consumption in the postanesthesia care unit (PACU) in a “real-world” clinical setting.

Study Design

This hospital registry study analyzed anesthesia cases taking place between January 2015 and June 2022 at Beth Israel Deaconess Medical Center, a tertiary healthcare center in Boston, Massachusetts. Ethical approval for the use of deidentified perioperative data from the institutional Anesthesia Research Data Repository was obtained from the institutional review board at Beth Israel Deaconess Medical Center (protocol No. 2023P000332). The requirement for informed consent was waived. Details on data sources and data collection are described in Supplemental Digital Content 1, section S1.1 (https://links.lww.com/ALN/D550). This article adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.10 

Study Population

Adult patients undergoing ambulatory procedures under general anesthesia who were admitted to the PACU were eligible for inclusion in this study. Patients who did not receive any opioids in the PACU were not excluded. The primary analysis was conducted using a complete-case method, and observations with missing data for a priori–defined confounding variables were excluded.

Exposure and Outcomes Measures

The primary exposure was defined as administration of acetaminophen either preoperatively in the holding area via the oral or intravenous route, or intraoperatively before extubation via the intravenous route. The potential effect modifier was defined as intraoperative administration of a 5-HT3 antagonist, including ondansetron, granisetron, dolasetron, palonosetron, or alosetron, via the intravenous route before extubation. Both exposure variables were binary (administration yes/no). The primary outcome was the cumulative dose of opioids administered in the PACU, quantified as oral morphine equivalents (OME) in milligrams.11 

Primary Analysis

To test our hypothesis that the coadministration of a 5-HT3 antagonist impacts the analgesic effect of acetaminophen, we included an interaction term between the administration of acetaminophen and administration of a 5-HT3 antagonist in the primary model. A multivariable, zero-inflated negative binomial regression model was applied. Analyses were adjusted for a priori–defined potential confounding variables selected based on existing literature and clinical relevance. These included patient demographics (age, sex, body mass index, federal insurance, estimated household income) and pre-existing comorbidities and risk indices (American Society of Anesthesiologists Physical Status classification, Elixhauser Comorbidity Index,12  Apfel score for postoperative nausea and vomiting,13  opioid use 90 days before,14,15  and history of chronic pain, depression,14  anxiety,14  drug abuse,14  liver disease, and renal failure). Moreover, preoperatively available variables (emergency status, administration of gabapentinoids or celecoxib in the holding area16,17 ) as well as factors related to intraoperative events, anesthesia, and the procedure itself (type and duration of procedure, work relative value units,18  use of regional anesthesia,19  intraoperatively administered dose of short- and long-acting opioids quantified as milligrams OME, crystalloid and colloid fluid infusion volume,20  number of packed erythrocyte units,21  age-adjusted minimum alveolar concentration of inhalational anesthetics, use of nitrous oxide,22  vasopressor agent,23  magnesium,24  nonsteroidal anti-inflammatory drugs, ketamine, dexmedetomidine,25  gabapentinoids, lidocaine, and use of antiemetics other than 5-HT3 receptor antagonists [dexamethasone,26  haloperidol, promethazine, scopolamine, metoclopramide, diphenhydramine], minutes of mean arterial pressure less than 55 mmHg23 ) were included. To account for potential changes in clinical practice over time, we further adjusted for the year in which the procedure was performed. Details on variable definitions and methodologic details to the primary model are provided in Supplemental Digital Content 1, sections S1.2 and S2 (https://links.lww.com/ALN/D550).

Secondary Analyses

In our secondary analyses, we assessed whether the effect of acetaminophen on postoperative opioid consumption was dose-dependent, taking into account the acetaminophen dose in milligrams per kilogram of body weight and investigated the potential interaction between 5-HT3 antagonists and acetaminophen dose. Furthermore, we tested if non–5-HT3 antagonist antiemetics influenced the analgesic effect of acetaminophen. Details on secondary analyses are provided in Supplemental Digital Content 1, section S3 (https://links.lww.com/ALN/D550).

Exploratory and Sensitivity Analyses

In exploratory analyses, we investigated the association between coadministration of acetaminophen and 5-HT3 antagonists with the length of stay in the PACU using a multivariable negative binomial regression. In addition, subgroup analysis was performed to examine whether the observed interaction differed between procedures that exceeded the median observed procedural duration compared to those that did not.

Moreover, we assessed for provider variability in the coadministration of 5-HT3 antagonists with acetaminophen by including individual anesthesia providers as a random effect in a mixed-effects logistic regression model. For this analysis, only cases in which acetaminophen was administered were considered, and the intraoperative coadministration of a 5-HT3 antagonist was set as the outcome to predict the probability of coadministration.

Further, we investigated the effects of acetaminophen on weight-adjusted dose of PACU opioids, measured in micrograms OME per kilogram body weight and separately when additionally adjusted for PACU length of stay (measured in micrograms OME per kilogram body weight per hour of PACU stay). Additionally, using the primary model, we tested whether the intraoperative administration of a 5-HT3 antagonist was associated with a change in PACU opioid doses. This analysis was additionally adjusted for acetaminophen use. Finally, we investigated how the timing of acetaminophen administration (pre- vs. intraoperative) affected the interaction with 5-HT3 antagonists and analgesic efficacy of acetaminophen.

Details on exploratory analyses are provided in Supplemental Digital Content 1, section S4 (https://links.lww.com/ALN/D550).

Multiple sensitivity analyses were conducted to test the robustness of our findings, including (1) multiple imputation of missing data, such as body mass index or estimated household income (fig. 1) and inclusion of patients who were previously excluded for missing data, (2) propensity score matching, and (3) E value calculation to test the robustness of the observed associations and assess potential unmeasured confounding in our analysis. Further, we (4) tested for effect modification of the association between acetaminophen and the dose of opioids administered in the PACU by investigating the association between acetaminophen administration and PACU opioid dose in subgroups of patients with and without 5-HT3 antagonists, (5) excluded patients who underwent surgery before 2017 to account for the lower incidence of opioid administration in our data in 2015 and 2016, and reassessed our primary analysis after excluding patients who received additional pain medication, such as (6) intraoperative nonsteroidal anti-inflammatory drugs or (7) pre- or intraoperative gabapentinoids. Moreover, we excluded patients who (8) did not receive any opioids in the PACU, (9) received tramadol in the PACU, (10) received dolasetron intraoperatively, or (11) underwent a type of surgery that was performed fewer than 500 times in the cohort during the study period, such as dental, ophthalmologic, or podiatric surgery. Last, we performed additional confounder adjustment, including (12) the preoperative use of antidepressants and (13) marital status as covariates in the primary model. Details on the conducted sensitivity analyses are provided in Supplemental Digital Content 1, section S5 (https://links.lww.com/ALN/D550).

Fig. 1.

Study flow diagram indicating numbers of patients screened and excluded due to missing data (Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts). *Multiple criteria may apply.

Fig. 1.

Study flow diagram indicating numbers of patients screened and excluded due to missing data (Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts). *Multiple criteria may apply.

Close modal

Statistical Analysis

Data are presented as median (with interquartile range) or frequency (with percentage) unless otherwise specified. Primary and secondary statistical analyses were conducted using prespecified outcomes and methods. We applied a multivariable, zero-inflated negative binomial regression model for the primary and secondary analyses. This model consists of a combination of a negative binomial model and a logistic model and was chosen because it allows accounting for overdispersion of the data (i.e., the variance of the data being much higher than their mean), while controlling for the covariates. A zero-inflated model was applied to account for the high number of patients who did not receive any postoperative opioids (OME dose = 0 mg) in the PACU, allowing for the estimation of the likelihood of no opioid requirement under various conditions.

Exploratory analyses included multivariable zero-inflated negative binomial regressions, multivariable zero-inflated negative binomial regressions, and multivariable mixed-effect logistic regressions. Continuous variables that did not have a linear relationship with the outcome were categorized into quantiles or clinically meaningful categories. Unless otherwise specified, results from analyses are reported as adjusted absolute differences with corresponding 95% CI. All analyses were performed using Stata (Version SE 16.1, StataCorp LLC, USA) and R Statistical Software (Version 4.3.0, available at: https://cran.r-project.org/bin/windows/base/old/4.3.0/; accessed November 1, 2022).

Study Cohort and Characteristics

We included 60,236 adult patients who underwent ambulatory procedures between 2015 and 2022. After exclusion of patients with missing data, the final study cohort consisted of 55,016 patients (fig. 1). Baseline characteristics of the study population are provided in table 1. A total of 3,166 of 55,016 patients (5.8%) received only acetaminophen, 15,438 of 55,016 patients (28.1%) received only 5-HT3 antagonists, 31,850 of 55,016 patients (57.9%) received both, and 4,562 of 55,016 patients (8.3%) received neither. A total of 47,278 of 47,288 (99.98%) of the patients receiving 5-HT3 antagonists received ondansetron, and 10 of 47,288 patients (0.02%) received dolasetron. The median acetaminophen dose was 1,000 mg (interquartile range, 1,000 to 1,000 mg) among 35,016 of 55,016 patients (63.6%) who received acetaminophen, with a mean ± SD acetaminophen dose of 638.5 ± 490.0 mg across all patients. The median doses of ondansetron and dolasetron were 4 mg (interquartile range, 4 to 4 mg) among 47,288 of 55,016 patients (86.0%) who received a 5-HT3 antagonist, with a mean ± SD dose of 3.5 ± 1.5 mg across all patients. A total of 16,640 of 55,016 patients (30.2%) received opioids in the PACU with a median dose of 7.5 mg OME (interquartile range, 7.5 to 14.3 mg OME). When including patients who did not receive opioids, the mean ± SD dose of PACU opioids was 3.2 ± 5.5 mg OME with a maximum of 20.4 mg OME. Data on PACU length of stay were available for 55,007 of 55,016 patients (more than 99.9%). The mean ± SD and median duration of PACU length of stay were 167.3 ± 90.9 min and 147.0 min (interquartile range, 114.0 to 195.0 min), respectively.

Table 1.

Patient and Procedural Characteristics by Levels of Interaction Term

Patient and Procedural Characteristics by Levels of Interaction Term
Patient and Procedural Characteristics by Levels of Interaction Term

Primary Analysis

In unadjusted analysis, acetaminophen administration was associated with higher opioid doses administered in the PACU (absolute difference, 1.70 mg OME; 95% CI, 1.55 to 1.85; P < 0.001). After adjustment for confounding variables, among all patients, there was no difference in postoperative opioid consumption between patients with and without acetaminophen administration (adjusted absolute difference, –0.02 mg OME; 95% CI, –0.08 to 0.03; P = 0.39).

The effect of acetaminophen on PACU opioid dose was modified by the coadministration of a 5-HT3 antagonist (P for interaction = 0.013): among patients who did not receive 5-HT3 antagonists, the administration of acetaminophen was associated with a –5.5% lower dose of opioids in the PACU (95% CI, –9.6 to –1.4%; P = 0.009; fig. 2). By contrast, among patients who received a 5-HT3 antagonist, the administration of acetaminophen was not associated with differences in opioid consumption in the PACU (–0.1%; 95% CI, –1.7 to 1.6%; P = 0.95; fig. 2). Figure 2 displays estimated (adjusted) opioid doses for patients who received and patients who did not receive acetaminophen with or without concomitant administration of a 5-HT3 antagonist.

Fig. 2.

Differential effects of acetaminophen on postoperative opioid requirements by the coadministration of a 5-hydroxytryptamine type 3 (5-HT3) antagonist. Estimated changes in opioid doses administered in the postanesthesia care unit, compared to no acetaminophen, among patients who did or did not receive a 5-HT3 antagonist intraoperatively and estimated average opioid doses administered in the postanesthesia care unit for different levels of the exposure. These estimates were adjusted for multiple covariates, including patient and procedural characteristics. Adjusted estimates are presented with corresponding 95% CIs. OME, oral morphine equivalents.

Fig. 2.

Differential effects of acetaminophen on postoperative opioid requirements by the coadministration of a 5-hydroxytryptamine type 3 (5-HT3) antagonist. Estimated changes in opioid doses administered in the postanesthesia care unit, compared to no acetaminophen, among patients who did or did not receive a 5-HT3 antagonist intraoperatively and estimated average opioid doses administered in the postanesthesia care unit for different levels of the exposure. These estimates were adjusted for multiple covariates, including patient and procedural characteristics. Adjusted estimates are presented with corresponding 95% CIs. OME, oral morphine equivalents.

Close modal

Secondary Analyses

The weight adjusted median dose of acetaminophen was 12.9 mg/kg (interquartile range, 11.0 to 15.4 mg/kg). An increase in acetaminophen dose was associated with a decrease in PACU opioid dose only when given alone and not in conjunction with a 5-HT3 antagonist (P for interaction = 0.004). Specifically, for every 1 mg/kg increase in acetaminophen, there was a reduction of –0.02 mg OME in PACU opioid dose (95% CI, –0.03 to –0.01; P = 0.002) in the absence of a 5-HT3 antagonist. By contrast, when a 5-HT3 antagonist was added, an increase in acetaminophen dose was no longer associated with a change in PACU opioid dose (0.00 mg OME; 95% CI, 0.00 to 0.00; P = 0.78). This relationship was further corroborated when using PACU opioids expressed as micrograms OME per kilogram (Supplemental Digital Content 1, section S3.1, https://links.lww.com/ALN/D550).

A total of 34,022 of 55,016 patients (61.8%) also received antiemetics other than 5-HT3 antagonists, either in addition to or instead of a 5-HT3 antagonist. Dexamethasone (32,682 of 55,016 [59.4%]), haloperidol (5,956 of 55,016 [10.8%]), and scopolamine (875 of 55,016 [1.6%]) were the most frequently administered non–5-HT3 antagonist antiemetics. There was no interaction between acetaminophen and intraoperative use of non–5-HT3 antagonist antiemetics (P for interaction = 0.49 for the composite exposure including all non–5-HT3 antagonist antiemetics; for results related to individual drugs, see Supplemental Digital Content 1, section S3.2, https://links.lww.com/ALN/D550).

Exploratory and Sensitivity Analyses

In a subgroup of 27,197 patients who underwent procedures that exceeded the median intervention duration (more than 97 min), acetaminophen was not associated with a change in PACU opioid dose (adjusted absolute difference, –0.02; 95% CI, –0.11 to 0.07; P = 0.71) and did not show an interaction with 5-HT3 antagonist coadministration (P for interaction = 0.20). By contrast, among 27,818 patients undergoing shorter procedures (97 min or less), the prophylactic use of acetaminophen was associated with a reduction in PACU opioid dose when administered without a 5-HT3 antagonist (adjusted absolute difference, –0.15; 95% CI, –0.29 to –0.02; P = 0.024 without 5-HT3 antagonist; P for interaction = 0.037). When investigating microgram OME per kilogram body weight and microgram OME per kilogram body weight per hour of PACU stay as alternative outcomes, we observed a similar interaction between acetaminophen and 5-HT3 antagonists, such that the analgesic effect of acetaminophen was eliminated when a 5-HT3 antagonist was coadministered (P for interaction = 0.002 and 0.028, respectively). There was no association between intraoperative 5-HT3 antagonist administration and PACU opioid dose (P = 0.41). For both pre- and intraoperative administration of acetaminophen, there was an interaction with 5-HT3 antagonist administration (P for interaction = 0.033 and < 0.001, respectively). We investigated the differing practices among our providers that had performed at least 50 included cases. There was a large variability in the incidence with which individual providers were predicted to coadminister acetaminophen and 5-HT3 antagonists, ranging from quite infrequently (2.1% of one provider’s cases) to very frequently (80.2% of another provider’s cases; fig. 3).

Fig. 3.

Anesthesia provider variability in the coadministration of 5-hydroxytryptamine type 3 (5-HT3) antagonists when using acetaminophen. The range of predicted probabilities for the coadministration of 5-HT3 antagonists (mean and 95% CI) in addition to acetaminophen across individual anesthesia providers was obtained from a mixed effect model including the confounders used in the primary regression model.

Fig. 3.

Anesthesia provider variability in the coadministration of 5-hydroxytryptamine type 3 (5-HT3) antagonists when using acetaminophen. The range of predicted probabilities for the coadministration of 5-HT3 antagonists (mean and 95% CI) in addition to acetaminophen across individual anesthesia providers was obtained from a mixed effect model including the confounders used in the primary regression model.

Close modal

Overall, pre- or intraoperative administration of acetaminophen was associated with a shorter PACU length of stay (adjusted absolute difference, –2.4 min; 95% CI, –3.8 to –0.9; P = 0.001), which was not modified by the coadministration of a 5-HT3 antagonist (P for interaction = 0.34).

Results of the primary analysis remained robust throughout all sensitivity analyses, which are detailed in Supplemental Digital Content 1, section S5 (https://links.lww.com/ALN/D550).

Power Statement

Statistical power was calculated on the basis of previously published data,27,28  where postoperative oxycodone consumption was reported as 0.50 mg/kg after perioperative administration of placebo or acetaminophen with ondansetron, and 0.35 mg/kg after perioperative administration of acetaminophen alone, with a pooled SD of 0.20 mg/kg, yielding an effect size of Cohen’s d = 0.75.27,28  The significance level α was set to 0.05. To be conservative, we used the sample size of the smallest exposure group (3,166 of 55,016 patients [5.8%] who received acetaminophen but no 5-HT3 antagonist) for the power calculation, assuming an equal sample size for exposure and control (no ondansetron, no acetaminophen).

Power was calculated to be greater than 0.99, indicating that the study possesses a high probability of detecting a true effect, thus minimizing the risk of a type II error. Power analysis was conducted as a two-sided two-sample t test using G*Power (version 3.1.9.4, Heinrich-Heine-University Duesseldorf, Germany).29 

In this hospital registry study of 55,016 patients who underwent general anesthesia for ambulatory procedures, the administration of pre- or intraoperative acetaminophen without intraoperative 5-HT3 antagonists was associated with a small and dose-dependent reduction in PACU opioid consumption. This association was not observed when a 5-HT3 antagonist was administered during anesthesia. This potential modification of the analgesic effect of acetaminophen was found exclusively with 5-HT3 antagonists and not with other commonly used antiemetics, such as dexamethasone or haloperidol. Prophylactic acetaminophen was not associated with lower opioid administration among patients undergoing longer procedures (more than 97 min), and therefore no interaction was observed in these patients.

Reduction of the analgesic effect of acetaminophen when 5-HT3 antagonists were coadministrated has previously been described in animal models and healthy volunteers.5–8,30  The clinical relevance of these findings was subsequently investigated by two small, randomized trials in children undergoing tonsillectomy. Ramirez et al. observed a threefold increase in postoperative opioid requirements when acetaminophen was administered in conjunction with a 5-HT3 antagonist.31  Similarly, Nikooseresht et al. reported that compared to acetaminophen alone, the combination of acetaminophen and a 5-HT3 antagonist resulted in a 96% increase in postoperative pain scores in the recovery unit.32  Furthermore, a randomized controlled trial by Koyuncu et al. involving 120 adult patients undergoing hysterectomy observed higher early postoperative pain scores when acetaminophen was administered in combination with a 5-HT3 antagonist at the end of surgery.33  While the current study does not test causal mechanisms, and the observed associations cannot prove that these mechanisms are pivotal here, it suggests potential relevance of preclinical findings by demonstrating an interaction between ondansetron and acetaminophen in a profoundly greater number of adult patients undergoing a variety of procedures. In addition, because all patients in the aforementioned trials received acetaminophen, these previous studies are unable to answer the question of what proportion of the effect of acetaminophen was eliminated by 5-HT3 antagonists. The design of our study allowed for estimation of the analgesic effect of acetaminophen with and without a 5-HT3 antagonist. Therefore, the current study is the first to show a complete absence of association between acetaminophen and postoperative opioid consumption when 5-HT3 antagonists were coadministered in a clinical patient collective.

Our observations may further elucidate seemingly conflicting results from previous studies. Jokela et al. reported no effect modification when patients received both 1,000 mg acetaminophen and 4 mg ondansetron in a cohort of 134 patients undergoing hysterectomy.28  Similar to the majority of patients in our cohort, patients in the study by Jokela et al. were administered acetaminophen at the beginning of surgery and ondansetron at the end. The median duration of surgery in the study by Jokela et al. was 134 min. This exceeds the 97-min threshold beyond which we no longer observed an association between acetaminophen administration and postoperative opioid consumption, even in patients without ondansetron administration, and also exceeds acetaminophen’s plasma half-life. Moreover, in the study by Jokela et al., acetaminophen administration was repeated every 6 h, thereby potentially exerting effects after the short plasma half-life of ondansetron (2.5 to 5.5 h).34  Finally, a randomized, double-blind trial including 32 patients undergoing ear, nose, and throat surgery under local anesthesia observed an increase in the analgesic effect of acetaminophen when a 5-HT3 antagonist was coadministered.35  However, due to a smaller than anticipated effect size (anticipated Cohen’s d [d] = 1; observed d = 0.47), this study might be underpowered, increasing the probability of false-negative results.

Although acetaminophen is among the most commonly used nonopioid analgesics worldwide and a first-line treatment option for mild to moderate acute pain,36  its mechanism of action remains poorly understood to date.4  Onset of acetaminophen analgesia is dependent on the route of administration with an average onset time of 3 to 5 min for intravenous administration and 11 min with a bioavailability of 70 to 90% if administered orally.37,38  Acetaminophen is distributed evenly across most body tissues.37  It is mostly metabolized in the liver, with a small fraction becoming a reactive metabolite neutralized by reduced glutathione before excretion.39  The drug’s plasma half-life is 1.9 to 2.5 h, with total body clearance of 4.5 to 5.5 ml · kg–1 · min–1.40  If administered in therapeutic doses, between 85 and 95% is excreted in the urine within 24 h, primarily as various conjugates.41  The effects of acetaminophen have been associated with central inhibition of prostaglandin synthesis,42  action on the endocannabinoid system,43  activation of the transient receptor potential vanilloid 1 and ankyrin 1,44,45  and effects on spinal 5-hydroxytryptamine pathways.5  The serotonergic effect of acetaminophen is presumably conveyed by an indirect activation of spinal serotonergic receptors, which in turn reinforce descending inhibitory pain pathways.5,6  While 5-HT3 antagonists primarily act on the chemoreceptor trigger zone in the medulla oblongata, they also inhibit spinal 5-HT3 receptors.5,46  With opposing effects on spinal 5-HT3 receptors, an agonist–antagonist interaction between the two drugs investigated in this study is likely and corroborated by our results. Our findings provide support based on clinical data for the notion that the analgesic effects of acetaminophen may be primarily attributed to the activation of serotonin receptors, which should further be elucidated in future studies.

Both optimal pain management and prevention of postoperative nausea and vomiting are crucial for patient satisfaction and recovery after surgery.3  Among prophylactic antiemetics, 5-HT3 antagonists are among the primary choices for many physicians due to their favorable risk/benefit profile.47  Our study provides important information for clinicians when selecting agents for prophylaxis of postoperative pain as well as postoperative nausea and vomiting. Based on our findings, one approach would be to administer acetaminophen as pain prophylaxis in shorter surgeries of less than 100 min. In these patients, ondansetron or other 5-HT3 antagonists may be reserved as rescue medication in order to avoid compromising the analgesic effects of acetaminophen. Antiemetic prophylaxes targeting other receptors, such as dexamethasone, dopamine antagonists, or neurokinin-1 receptor antagonists, may instead be considered.48  It is important to note that the majority of patients in our study received acetaminophen 1,000 mg orally as prophylaxis before surgery in the holding area and that the effect size of acetaminophen administration was small. Future studies are required to examine the interaction with 5-HT3 antagonists when acetaminophen is administered intravenously at the end of the procedure, which, according to our data, may be preferred over preoperative administration for longer surgeries of greater than 100 minutes. The observed variability in the coadministration of 5-HT3 antagonists among anesthesia providers who used acetaminophen in our cohort emphasizes the need for standardized protocols outlining which clinical scenarios require 5-HT3 antagonists or acetaminophen.49  Finally, our study provides important information for future studies investigating the analgesic effect of acetaminophen, where 5-HT3 antagonists might be avoided, or their administration thoroughly protocolized.

There are several limitations to this study. First, due to the retrospective design, residual confounding cannot be ruled out. However, the calculated E value of 1.33 for the point estimate indicates that an unmeasured confounder’s minimum strength of association with both the exposure and outcome would need to be 1.33 on the risk ratio scale, after accounting for the comprehensive confounder model, to explain away the observed exposure–outcome association. This suggests that the observed results are relatively robust against potential unaccounted confounding. Nonetheless, this study is limited to the availability of observational data recorded during clinical routine. For example, preoperative pain levels and the individual level of opioid tolerance for each patient, potentially important confounding variables, were not recorded during clinical routine and therefore not available in our dataset. We addressed this by including diagnoses of chronic pain conditions50  and data on preoperative opioid prescriptions as covariates. Second, as our study used electronic medical records, some data may have been incomplete or inaccurate. We addressed potential bias from years with unexpectedly low incidences of opioid administration in the PACU by excluding the respective period in a repeated primary analysis, which yielded robust results. Despite the large and diverse study cohort, including a variety of surgical services and procedures, a single study from one academic medical center may limit the generalizability to other populations, procedures, or healthcare settings. Furthermore, the lack of heterogeneity among different types of 5-HT3 antagonists examined in our study raises the possibility that our findings may not directly translate to other 5-HT3 antagonists. However, as ondansetron is the most commonly used 5-HT3 antagonist in the United States,51  we believe that the results from our study are clinically relevant.

In summary, an association between acetaminophen and lower opioid consumption in the PACU was not observed when 5-HT3 antagonists were coadministered. Based on our findings, physicians should consider alternative pain prophylaxis strategies in patients who receive 5-HT3 antagonists for prophylaxis of postoperative nausea and vomiting. Conversely, where acetaminophen is favored for pain prophylaxis, 5-HT3 antagonists can be reserved as rescue medication.

Acknowledgments

The authors are grateful to Max N. Hentges, B.Sc. (Department of Anesthesia, Critical Care and Pain Medicine, and Center for Anesthesia Research Excellence, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts), for his assistance in revising the text and language of the manuscript.

Research Support

Support was provided solely from institutional and/or departmental sources.

Competing Interests

Dr. Schaefer received funding for investigator-initiated studies from Merck & Co. (Rahway, New Jersey), which do not pertain to this article. He is an associate editor for BMC Anesthesiology. He received honoraria for lectures from Mindray Medical International Limited (Shenzen, China). He received an unrestricted philanthropic grant from Jeffrey and Judith Buzen. The other authors declare no competing interests.

Supplemental Digital Content 1, https://links.lww.com/ALN/D550

Supplemental Digital Content 2, https://links.lww.com/ALN/D551

STROBE Checklist, https://links.lww.com/ALN/D552

1.
Agrawal
S
,
Khazaeni
B
:
Acetaminophen toxicity, StatPearls.
StatPearls Publishing
,
2023
. . Accessed August 22, 2023.
2.
Acetaminophen market.
.
3.
Simpson
JC
,
Bao
X
,
Agarwala
A
:
Pain management in Enhanced Recovery After Surgery (ERAS) protocols.
Clin Colon Rectal Surg
2019
;
32
:
121
8
4.
Anderson
BJ
:
Paracetamol (acetaminophen): mechanisms of action.
Paediatr Anaesth
2008
;
18
:
915
21
5.
Pickering
G
,
Loriot
M-A
,
Libert
F
,
Eschalier
A
,
Beaune
P
,
Dubray
C
:
Analgesic effect of acetaminophen in humans: First evidence of a central serotonergic mechanism.
Clin Pharmacol Ther
2006
;
79
:
371
8
6.
Pelissier
T
,
Alloui
A
,
Caussade
F
et al
:
Paracetamol exerts a spinal antinociceptive effect involving an indirect interaction with 5-hydroxytryptamine3 receptors: In vivo and in vitro evidence.
J Pharmacol Exp Ther
1996
;
278
:
8
14
7.
Alloui
A
,
Pelissier
T
,
Dubray
C
,
Lavarenne
J
,
Eschalier
A
:
Tropisetron inhibits the antinociceptive effect of intrathecally administered paracetamol and serotonin.
Fundam Clin Pharmacol
1996
;
10
:
406
7
8.
Bandschapp
O
,
Filitz
J
,
Urwyler
A
,
Koppert
W
,
Ruppen
W
:
Tropisetron blocks analgesic action of acetaminophen: A human pain model study.
Pain
2011
;
152
:
1304
10
9.
Weiser
TG
,
Haynes
AB
,
Molina
G
et al
:
Size and distribution of the global volume of surgery in 2012.
Bull World Health Organ
2016
;
94
:
201F
9F
10.
von Elm
E
,
Altman
DG
,
Egger
M
,
Pocock
SJ
,
Gøtzsche
PC
,
Vandenbroucke
JP
;
STROBE Initiative
:
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Guidelines for reporting observational studies.
PLoS Med
2007
;
4
:
e296
11.
Nielsen
S
,
Degenhardt
L
,
Hoban
B
,
Gisev
N
:
A synthesis of oral morphine equivalents (OME) for opioid utilisation studies.
Pharmacoepidemiol Drug Saf
2016
;
25
:
733
7
12.
Walraven
C van
,
Austin
PC
,
Jennings
A
,
Quan
H
,
Forster
AJ
:
A modification of the Elixhauser comorbidity measures into a point system for hospital death using administrative data.
Med Care
2009
;
47
:
626
33
13.
Apfel
CC
,
Läärä
E
,
Koivuranta
M
,
Greim
CA
,
Roewer
N
:
A simplified risk score for predicting postoperative nausea and vomiting: conclusions from cross-validations between two centers.
Anesthesiology
1999
;
91
:
693
700
14.
Armstrong
AD
,
Hassenbein
SE
,
Black
S
,
Hollenbeak
CS
;
Interdisciplinary Pain Team
:
Risk factors for increased postoperative pain and recommended orderset for postoperative analgesic usage.
Clin J Pain
2020
;
36
:
845
51
15.
Kuck
K
,
Naik
BI
,
Domino
KB
et al
;
Multicenter Perioperative Outcomes Group Enhanced Observation Study Investigator Group for the Multicenter Perioperative Outcomes Group Enhanced Observation Study Collaborator Group
:
Prolonged opioid use and pain outcome and associated factors after surgery under general anesthesia: A prospective cohort association multicenter study.
Anesthesiology
2023
;
138
:
462
76
16.
Dirks
J
,
Fredensborg
BB
,
Christensen
D
,
Fomsgaard
JS
,
Flyger
H
,
Dahl
JB
:
A randomized study of the effects of single-dose gabapentin versus placebo on postoperative pain and morphine consumption after mastectomy.
Anesthesiology
2002
;
97
:
560
4
17.
Wachtendorf
LJ
,
Schaefer
MS
,
Santer
P
et al
:
Association between preoperative administration of gabapentinoids and 30-day hospital readmission: A retrospective hospital registry study.
J Clin Anesth
2021
;
73
:
110376
18.
Santer
P
,
Wachtendorf
LJ
,
Suleiman
A
et al
:
Mechanical power during general anesthesia and postoperative respiratory failure: A multicenter retrospective cohort study.
Anesthesiology
2022
;
137
:
41
54
19.
Polshin
V
,
Petro
J
,
Wachtendorf
LJ
et al
:
Effect of peripheral nerve blocks on postanesthesia care unit length of stay in patients undergoing ambulatory surgery: A retrospective cohort study.
Reg Anesth Pain Med
2021
;
46
:
233
9
20.
Shin
CH
,
Long
DR
,
McLean
D
et al
:
Effects of intraoperative fluid management on postoperative outcomes: A hospital registry study.
Ann Surg
2018
;
267
:
1084
92
21.
Althoff
FC
,
Wachtendorf
LJ
,
Rostin
P
et al
:
Effects of night surgery on postoperative mortality and morbidity: A multicentre cohort study.
BMJ Qual Saf
2021
;
30
:
678
88
22.
Sanders
RD
,
Weimann
J
,
Maze
M
:
Biologic effects of nitrous oxide: A mechanistic and toxicologic review.
Anesthesiology
2008
;
109
:
707
22
23.
Wachtendorf
LJ
,
Azimaraghi
O
,
Santer
P
et al
:
Association between intraoperative arterial hypotension and postoperative delirium after noncardiac surgery: A retrospective multicenter cohort study.
Anesth Analg
2022
;
134
:
822
33
24.
De Oliveira
GS
,
Castro-Alves
LJ
,
Khan
JH
,
McCarthy
RJ
:
Perioperative systemic magnesium to minimize postoperative pain: A meta-analysis of randomized controlled trials.
Anesthesiology
2013
;
119
:
178
90
25.
Ma
H
,
Wachtendorf
LJ
,
Santer
P
et al
:
The effect of intraoperative dexmedetomidine administration on length of stay in the post-anesthesia care unit in ambulatory surgery: A hospital registry study.
J Clin Anesth
2021
;
72
:
110284
26.
Blank
M
,
Katsiampoura
A
,
Wachtendorf
LJ
et al
:
Association between intraoperative dexamethasone and postoperative mortality in patients undergoing oncologic surgery: A multicentric cohort study.
Ann Surg
2023
;
278
:
e105
14
27.
Jokela
R
,
Ahonen
J
,
Tallgren
M
,
Haanpää
M
,
Korttila
K
:
A randomized controlled trial of perioperative administration of pregabalin for pain after laparoscopic hysterectomy.
Pain
2008
;
134
:
106
12
28.
Jokela
R
,
Ahonen
J
,
Seitsonen
E
,
Marjakangas
P
,
Korttila
K
:
The influence of ondansetron on the analgesic effect of acetaminophen after laparoscopic hysterectomy.
Clin Pharmacol Ther
2010
;
87
:
672
8
29.
Faul
F
,
Erdfelder
E
,
Lang
A-G
,
Buchner
A
:
G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences.
Behav Res Methods
2007
;
39
:
175
91
30.
Tiippana
E
,
Hamunen
K
,
Kontinen
V
,
Kalso
E
:
The effect of paracetamol and tropisetron on pain: Experimental studies and a review of published data.
Basic Clin Pharmacol Toxicol
2013
;
112
:
124
31
31.
Ramirez
L
,
Cros
J
,
Marin
B
et al
:
Analgesic interaction between ondansetron and acetaminophen after tonsillectomy in children: The Paratron randomized, controlled trial.
Eur J Pain
2015
;
19
:
661
8
32.
Nikooseresht
M
,
Nasrolahi
M
,
Hajian
P
,
Moradi
A
:
The effects of ondansetron on the analgesic action of intravenous acetaminophen after tonsillectomy in children: A triple-blind randomized controlled trial.
Oxid Med Cell Longev
2021
;
2021
:
6611740
33.
Koyuncu
O
,
Leung
S
,
You
J
et al
:
The effect of ondansetron on analgesic efficacy of acetaminophen after hysterectomy: A randomized double blinded placebo controlled trial.
J Clin Anesth
2017
;
40
:
78
83
34.
Pritchard
JF
:
Ondansetron metabolism and pharmacokinetics.
Semin Oncol
1992
;
19
:
9
15
35.
Bhosale
UA
,
Khobragade
R
,
Naik
C
,
Yegnanarayan
R
,
Kale
J
:
Randomized, double-blind, placebo-controlled study to investigate the pharmacodynamic interaction of 5-HT3 antagonist ondansetron and paracetamol in postoperative patients operated in an ENT department under local anesthesia.
J Basic Clin Physiol Pharmacol
2015
;
26
:
217
22
36.
Anekar
AA
,
Cascella
M
:
WHO analgesic ladder, StatPearls.
StatPearls Publishing
,
2022
.
Available at: https://www.ncbi.nlm.nih.gov/books/NBK554435/. Accessed February 15, 2023
.
37.
Forrest
JA
,
Clements
JA
,
Prescott
LF
:
Clinical pharmacokinetics of paracetamol.
Clin Pharmacokinet
1982
;
7
:
93
107
38.
Moller
PL
,
Sindet-Pedersen
S
,
Petersen
CT
,
Juhl
GI
,
Dillenschneider
A
,
Skoglund
LA
:
Onset of acetaminophen analgesia: Comparison of oral and intravenous routes after third molar surgery.
Br J Anaesth
2005
;
94
:
642
8
39.
Adriaenssens
PI
,
Prescott
LF
:
High performance liquid chromatographic estimation of paracetamol metabolites in plasma.
Br J Clin Pharmacol
1978
;
6
:
87
8
40.
Arnman
R
,
Olsson
R
:
Elimination of paracetamol in chronic liver disease.
Acta Hepatogastroenterol (Stuttg)
1978
;
25
:
283
6
41.
Andrews
RS
,
Bond
CC
,
Burnett
J
,
Saunders
A
,
Watson
K
:
Isolation and identification of paracetamol metabolites.
J Int Med Res
1976
;
4
:
34
9
42.
Flower
RJ
,
Vane
JR
:
Inhibition of prostaglandin synthetase in brain explains the anti-pyretic activity of paracetamol (4-acetamidophenol).
Nature
1972
;
240
:
410
1
43.
Ottani
A
,
Leone
S
,
Sandrini
M
,
Ferrari
A
,
Bertolini
A
:
The analgesic activity of paracetamol is prevented by the blockade of cannabinoid CB1 receptors.
Eur J Pharmacol
2006
;
531
:
280
1
44.
Mallet
C
,
Barrière
DA
,
Ermund
A
et al
:
TRPV1 in brain is involved in acetaminophen-induced antinociception.
PLoS One
2010
;
5
:
e12748
45.
Andersson
DA
,
Gentry
C
,
Alenmyr
L
et al
:
TRPA1 mediates spinal antinociception induced by acetaminophen and the cannabinoid Δ(9)-tetrahydrocannabiorcol.
Nat Commun
2011
;
2
:
551
46.
Tyers
MB
,
Bunce
KT
,
Humphrey
PP
:
Pharmacological and anti-emetic properties of ondansetron.
Eur J Cancer Clin Oncol
1989
;
25
(
suppl 1
):
S15
9
47.
Weibel
S
,
Rücker
G
,
Eberhart
LH
et al
:
Drugs for preventing postoperative nausea and vomiting in adults after general anaesthesia: A network meta-analysis.
Cochrane Database Syst Rev
2020
;
2020
:
CD012859
48.
Gan
TJ
,
Belani
KG
,
Bergese
S
et al
:
Fourth consensus guidelines for the management of postoperative nausea and vomiting.
Anesth Analg
2020
;
131
:
411
48
49.
White
RS
,
Andreae
MH
,
Lui
B
et al
;
Multicenter Perioperative Outcomes Group Collaborators
:
Antiemetic administration and its association with race: A retrospective cohort study.
Anesthesiology
2023
;
138
:
587
601
50.
Foley
HE
,
Knight
JC
,
Ploughman
M
,
Asghari
S
,
Audas
R
:
Association of chronic pain with comorbidities and health care utilization: A retrospective cohort study using health administrative data.
Pain
2021
;
162
:
2737
49
51.
Tricco
AC
,
Soobiah
C
,
Blondal
E
et al
:
Comparative efficacy of serotonin (5-HT3) receptor antagonists in patients undergoing surgery: A systematic review and network meta-analysis.
BMC Med
2015
;
13
:
136