Assessment of Opioid Use and Analgesic Requirements After Endoscopic Sinus Surgery: A Randomized Clinical Trial

Key Points

Question  Will different analgesic regimens prescribed after endoscopic sinus surgery affect the degree of postoperative pain experienced and number of opioids consumed?

Findings  In this randomized, multi-institutional clinical trial, the 100 participants who were both randomized and analyzed consumed on average fewer than 2 opioid tablets after endoscopic sinus surgery. Opioid consumption and the degree of pain experienced did not notably change with the introduction of ibuprofen in 1 analgesic regimen.

Meaning  Most patients who undergo endoscopic sinus surgery can be prescribed fewer narcotics than the current norm, and clinicians can consider permitting ibuprofen use postoperatively.

Importance  The opioid epidemic has generated interest in optimizing opioid prescribing after common surgeries. Recent studies have shown a broad range of analgesic prescription patterns following endoscopic sinus surgery (ESS).

Objective  To compare the efficacy of different analgesic regimens after ESS.

Design, Setting, and Participants  This multi-institutional, nonblinded randomized clinical trial was conducted at 6 tertiary centers across the US and Canada and included participants who underwent ESS for acute or chronic rhinosinusitis. The study was conducted from March 2019 to March 2020, and the data were analyzed in November to December 2020.

Interventions  All participants received acetaminophen, 650 mg, as the first-line analgesic. From there, patients were randomized to either oxycodone rescue (oxycodone, 5 mg, as second-line therapy) or ibuprofen rescue (ibuprofen, 600 mg, as second-line therapy, with oxycodone, 5 mg, reserved for breakthrough pain).

Main Outcomes and Measures  Baseline characteristics and disease severity were collected at enrollment. Medication logs, pain scores, and epistaxis measures were collected until postoperative day 7. The primary outcome was the postoperative visual analog scale score for pain. Brief Pain Inventory Pain Severity and Pain Interference Scores were also collected.

Results  A total of 118 patients were randomized (62 [52.5%] oxycodone rescue, 56 [47.5%] ibuprofen rescue; mean [SD] age, 46.7 [16.3] years; 44 women [44.0%]; 83 White [83.0%], 7 Black [7.0%], and 7 Asian individuals [7.0%]). After exclusions for loss to follow-up and noncompliance, 51 remained in the oxycodone rescue group and 49 in the ibuprofen rescue group. The groups had similar demographic characteristics and disease severity. Thirty-two (63%) in the oxycodone rescue group had adequate pain management with acetaminophen only, while 19 (37%) consumed at least 1 oxycodone dose. In the ibuprofen rescue group, 18 (16%) required only acetaminophen, 28 (57%) used only acetaminophen and ibuprofen, and the remaining 13 (26%) consumed 1 or more oxycodone doses. The groups had similar average acetaminophen (9.69 vs 7.96 doses; difference, 1.73; 95% CI, –1.37 to 4.83) and oxycodone (1.89 vs 0.77 doses; difference, 1.13; 95% CI, –0.11 to 2.36) use. Both groups had similar postoperative visual analog scale scores. A subanalysis that compared opioids users with nonusers showed clinically significant lower pain scores in nonusers at multiple postoperative points.

Conclusions and Relevance  In this randomized clinical trial, most patients who underwent ESS could be treated postoperatively using a nonopioid regimen of either acetaminophen alone or acetaminophen and ibuprofen. Ibuprofen as a second-line therapy did not reduce overall narcotic consumption, but the overall narcotic use was low in both groups.

Trial Registration  ClinicalTrials.gov Identifier: NCT03783702

The opioid epidemic remains a national public health crisis, with high associated morbidity and mortality and extensive socioeconomic costs.^1,2 Deaths of drug overdoses tripled from 1999 to 2014 and continue to increase annually. In 2015, more than 33 000 deaths in the US were directly linked to opioid overdose, roughly 5 times the rate seen in 1999.^3,4 Sharing of unused opioids with friends and family members has become the single largest contributor to opioid abuse.^5-7

Any opioid consumption after surgery puts patients at risk for long-term narcotic use. A 2017 US Centers for Disease Control and Prevention study reported high rates of opioid dependence following prescriptions provided for short-term and long-term therapy, including among opioid-naive patients.⁴ Among patients who took 1 day of prescribed opioids, 6% continued to use opioids 1 year later, and 2.9% continued opioid use 3 years later. The prevalence of persistent narcotic use increased to 13.5% at 1 year when the first opioid course lasted for 8 or more days.

Within otolaryngology, rhinologic surgery offers an opportunity for assessing postsurgical opioid management. There are more than 250 000 endoscopic sinus surgeries (ESSs) and 260 000 septoplasties performed annually in the US.⁸ Multiple studies have shown that, for patients recovering from ESS, otolaryngologists typically prescribe quantities of opioids that exceed the amount patients actually consume.^9-12 A recent survey of American Rhinologic Society members showed that 94% of members prescribe opioids after ESS, with an average of 27.4 tablets prescribed to a patient per surgery.¹³ In contrast, an evidence-based review highlighted the growing evidence supporting the successful management of post-ESS pain using nonnarcotic analgesics.¹⁴

Although ESS is commonly performed throughout the US, to our knowledge, no standard postoperative analgesic regimen has been adopted or validated, and the existing literature reports wide ranging values for normative opioid consumption after these procedures. The purpose of this study is to better understand analgesic requirements after ESS and determine whether nonsteroidal anti-inflammatory drugs (NSAIDs) use can reduce narcotic requirements while maintaining adequate pain control.

This was a multi-institutional, prospective, nonblinded, randomized clinical trial (NCT03783702), and the trial protocol is provided in Supplement 1. Institutional review board approval was obtained at all institutions before enrollment. Participants were enrolled at 6 institutions: Stanford Health Care (Stanford, California), Mayo Clinic in Rochester (Minnesota), Vanderbilt Health (Nashville, Tennessee), Albert Einstein College of Medicine (Bronx, New York), NorthShore University Health System (Chicago, Illinois), and University of British Columbia (Vancouver, British Columbia, Canada). The study used the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.

Patients were eligible if they were 18 years or older and scheduled to undergo primary or revision ESS for chronic rhinosinusitis with nasal polyposis (CRSwNP), chronic rhinosinusitis without nasal polyposis (CRSsNP), or recurrent acute rhinosinusitis (RARS).

Participants were excluded if they had a history of chronic pain disorder, gastrointestinal ulcers or bleeding, chronic kidney disease or other known decreased kidney function (estimated glomerular filtration rate <60), liver cirrhosis or other liver impairment, prior adverse reactions to opioids or NSAIDs, alcohol or opioid use disorder, or other contraindications to any drug classes in either group. Patients taking an average of more than 4 tablets of NSAIDs or acetaminophen in a typical week or who took any narcotics or neuromodulating agents (eg, opioids, tramadol, nortriptyline, amitriptyline, gabapentin, pregabalin, or duloxetine) in the week before surgery were also excluded. Patients undergoing a concurrent Draf III frontal sinusotomy, septorhinoplasty, and/or expected to receive splints of any kind were also excluded.

All patients provided written informed consent in clinic. Patients were randomized into 1 of 2 treatment groups: oxycodone rescue or ibuprofen rescue. Randomization was performed using an online random number generator. Participants were sorted based on a random generation of odd or even numbers until there was a minimum of 55 participants per group. The study coordinators thoroughly explained the study protocol in a stepwise fashion and answered all questions. At the time of enrollment, participants were given a copy of their respective flow diagram, instructed to refer to it for stepwise management of postoperative pain, and asked to call the researchers with any questions about analgesic consumption (eFigure 1 in Supplement 2). Additional, study-specific counseling regarding opioid consumption outside of the routine protocol of clinicians was not performed.

Medications were prescribed at the time of enrollment or on the day of surgery to the patient’s preferred pharmacy. All were prescribed 15 tablets of oxycodone, 5 mg, and 50 tablets of acetaminophen, 650 mg. Those in the ibuprofen group were also prescribed ibuprofen, 600 mg. All were instructed to take one 650-mg acetaminophen tablet as the first-line agent for pain. Those in the oxycodone rescue group were instructed to take one 5-mg oxycodone dose as the second-line agent for pain if they felt their pain was insufficiently controlled with acetaminophen. In contrast, participants in the ibuprofen rescue group were instructed to take one 600-mg ibuprofen dose as the second-line agent for pain, with oxycodone, 5 mg, reserved as the third-line agent if participants felt they still needed additional pain control. All were asked to wait at least 30 to 45 minutes after taking 1 medication to ensure adequate time for effect onset before determining the need for additional analgesics. Participants were instructed not to divide the analgesic tablets into smaller doses.

All participants were instructed to initiate saline rinses on postoperative day (POD) 1. Patients with CRSwNP were prescribed a 12-day prednisone taper if no contraindications (eg, diabetes) existed.^15,16 Whereas for patients with CRSsNP or RARS, the decision regarding postoperative steroid use was left up to the attending surgeon. Postoperative continuation of any topical corticosteroid medications (eg, fluticasone spray or budesonide irrigation) was left up to surgeon preference. Antibiotic use was left to surgeon discretion.

Baseline and demographic characteristics, medical and surgical history, and a list of current medications was collected at enrollment. Comparison of preoperative disease severity was evaluated using the 22-item Sino-Nasal Outcome Test (SNOT-22) and Lund-Mackay scoring system.^17,18 Postoperatively, participants were asked to complete a daily medication log, and daily and cumulative analgesic consumption was tallied and compared. Patients were asked to record any deviations from the study protocol on the questionnaires and were queried about any deviations at their postoperative appointment.

Participants were asked to report pain scores nightly based on their average pain levels during the preceding 24 hours. The primary outcome of this study was the average daily visual analog scale (VAS) score for pain, which was recorded preoperatively and nightly for the first 7 PODs. The VAS is a continuous, patient-reported outcome measure that is determined using a horizontal 100-mm scale that ranges from no pain (score of 0) to worst imaginable pain (score of 100). Based on prior literature, a change of 13 mm was considered to be the minimal clinically important difference (MCID) for acute pain severity on the VAS.^19-21

The validated 24-hour Brief Pain Inventory (BPI) Severity and Interference Scale scores were collected at baseline and on each POD for 7 days. The BPI short form is a validated, patient-reported outcome measure that assesses pain over a 24-hour period.^22,23 The questionnaire is subdivided into the BPI Pain Severity and BPI Pain Interference Scores. The Pain Severity score is calculated as a composite mean score using the degree of pain a patient experiences per day at its least, worst, average, and now. The BPI Interference score determines the degree to which pain has interfered with daily life, including work, mood, general activity, walking, joy, sleep, and relations with others. Each score ranges from 0 to 10, with an MCID of 1.0. A consensus panel in 2003 recommended that the BPI be used for all chronic pain studies, and it is frequently used in acute pain clinical trials.²⁴

The degree of epistaxis was assessed using a horizontal 100-mm VAS scale, with participants asked to describe their bleeding on a continuum between “no bleeding to continuous bleeding. Participants also recorded the frequency and duration of daily nasal bleeding episodes.

The power calculation was based on the primary outcome of VAS score for pain. A 2-sided 2-sample equal variance t test was performed using the MCID of 13 mm for acute pain as the effect size, standard deviation for acute pain of 23.2, a significance level of .05, and power of 0.8. The minimum recommended sample size was determined to be 50 per arm.²⁵ With a total sample size of 100, there was an 80% chance of obtaining a statistically significant difference between the 2 treatment groups at a 5% significance level. To account for potential loss to follow-up and dropout, 118 patients were enrolled.

Statistical analysis was performed using SAS, version 9.4 (SAS Institute). The study outcomes are reported based on the per-protocol analysis. Baseline characteristics, comorbidities, the extent of sinus surgery, medication use, and outcome measures were compared across treatment groups, and effect size, along with 95% confidence intervals, was presented. The t test was used to compare the VAS, BPI severity, and BPI interference scores at baseline and each follow-up visit. Multivariable linear mixed models with repeated measures were performed to examine the concurrent associations between 7 repeated treatments and outcome measures.

A total of 118 patients were enrolled, 62 (52.5%) of whom were in the oxycodone rescue group and 56 (47.5%) in the ibuprofen rescue group (eFigure 2 in Supplement 2; Table). Eight (12.9%) in the oxycodone rescue group and 6 (10.7%) in the ibuprofen rescue group did not return the questionnaires and were considered lost to follow-up. In addition, 3 in the oxycodone rescue and 1 in the ibuprofen rescue group did not follow the protocol strictly. This left 51 in the oxycodone rescue and 49 in the ibuprofen rescue group with complete and fully accurate data. In all, 44 (44%) were women, with an overall mean (SD) age of 46 (16) years. Those in the ibuprofen rescue group were on average 6.9 years (95% CI, 0.6-13.3) older than those in the oxycodone rescue group. The groups had similar distributions for sex, race, body mass index (calculated as weight in kilograms divided by height in meters squared), baseline condition (acute or chronic rhinosinusitis), and comorbidities. They also had similar degrees of baseline sinonasal disease, as measured subjectively with preoperative SNOT-22 scores and objectively with Lund-Mackay scores. The extent of sinus surgery performed, in terms of the number and types of sinuses opened and the frequency of septoplasty and turbinate reduction, was similar between the 2 groups. Preoperatively, baseline pain scores were also similar in both groups, as measured using the VAS, BPI severity, and BPI interference scores.

The VAS scores were similar between both groups at baseline (mean [SD], 8.5 [4.9] for the oxycodone rescue group vs 12.39 [6.64] for the ibuprofen rescue group; difference, –3.8; 95% CI –10.6 to 2.9). Postoperatively, there were no clinically significant differences in pain scores detected. The VAS scores were similar across all postoperative points: POD 1 difference, 0.9 (95% CI, –9.6 to 11.4); POD 2 difference, 4.0 (95% CI, –5.5 to 13.5); POD 3 difference, –2.5 (95% CI, –10.9 to 5.9); POD 4 difference, –2.6 (95% CI, –9.9 to 4.8); POD 5 difference, 0.9 (95% CI, –7.6 to 9.3); POD 6 difference, 2.4 (95% CI, –4.6 to 9.4); and POD 7 difference, –2.0, (95% CI, –6.6 to 2.5) (Figure 1). The mean (SD) BPI severity scores were also similar at baseline (1.0 [1.5] for the oxycodone rescue group vs 1.6 [2.2] for the ibuprofen rescue group; difference, –0.6 (95% CI, –1.3 to 0.1) and during the 7 PODs recorded (POD 1 difference, 0; 95% CI, –0.9 to 0.8; POD 4 difference, –0.3; 95% CI, –1.0 to 0.4; POD 7 difference, 0; 95% CI, –0.7 to 0.7) (Figure 2). The BPI interference scores were lower in the ibuprofen rescue group on POD 4 (difference, –1.0; 95% CI, –1.9 to –0.2) but not at any other point (POD 1 difference, –0.1; 95% CI, –1.2 to 1.0; POD 3 difference, –0.8; 95% CI, –1.7 to 0.1; and POD 7 difference, –0.6; 95% CI, –1.5 to 0.3) (Figure 3). A multivariable linear mixed model with a repeated measure adjusting for the baseline level of outcome variable was performed, and the findings were similar. There was no statistically significant difference for VAS at any follow-up, but the BPI severity score was significantly different at POD 1 (difference, –0.7; 95% CI, –1.2 to –0.1), POD 3 (difference, –0.9; 95% CI, –1.6 to –0.1), and POD 4 (difference, –0.4; 95% CI, –1.8 to –0.2). The BPI interference scores also differed at POD 3 (difference, –1.2; 95% CI, –2.1 to –0.3), POD 4 (difference, –1.4; 95% CI, –2.4 to –0.5), POD 5 (difference, –1.1; 95% CI, –2.2 to –0.04), and POD 6 (difference, –1.1; 95% CI, –2.2 to –0.04). This portion of the analysis compares pain scores based on the randomization groups regardless of whether participants in each group consumed oxycodone or ibuprofen during the study period.

Post hoc subanalyses were performed with participants who consumed ibuprofen and/or oxycodone to better evaluate the importance of ibuprofen and oxycodone in pain reduction. Pain scores in participants who consumed any opioid (ie, 1 or more doses throughout the study period) were first compared with scores of those who did not use any opioids. When compared with those who consumed any opioid, the nonopioid users consistently had statistically and clinically significantly lower daily VAS (POD 1 difference, 22.5; 95% CI, 11.4 to 33.6); POD 5 difference, 27.9; 95% CI, 14.6 to 41.3; Figure 4), BPI severity (POD 1 difference, 1.6; 95% CI, 0.7 to 2.5; POD 5 difference, 2.3; 95% CI, 1.0 to 3.7; eFigure 3 in Supplement 2), and BPI interference (POD 1 difference, 1.9; 95% CI, 0.7 to 3.1; POD 5 difference, 3.5; 95% CI, 1.9 to 5.1; eFigure 4 in Supplement 2) scores at POD 1 to 5. See eResults 1 in Supplement 2 for more detailed analysis.

Aggregate and daily medication consumption for each prescription class was compared across both groups. In the oxycodone rescue group, 32 patients (63%) had their pain adequately managed with acetaminophen only while 19 (37%) consumed 1 or more 5-mg oxycodone dose. Eight patients in the ibuprofen rescue group (16%) had their pain adequately managed with only acetaminophen, 28 (57%) used only acetaminophen and ibuprofen, and the remaining 13 (26%) consumed 1 or more 5-mg oxycodone dose. Fewer than half of patients (19 [37%] in the oxycodone rescue group and 13 [26%] in the ibuprofen rescue group) consumed 1 or more 5-mg opioid doses over the 7-day period (difference, 12.7%; 95% CI, 0% to 34%). There was no difference in cumulative acetaminophen ingestion between the 2 groups over the study period. Participants in the oxycodone rescue group consumed a cumulative mean (SD) of 9.69 (7.89) acetaminophen doses over the 7-day period compared with 7.96 (SD 7.23) in the ibuprofen rescue group (difference, 1.73; 95% CI, –1.37 to 4.83). Comparing acetaminophen consumption between the groups by day showed no difference in daily consumption rates. The patients in the ibuprofen rescue group consumed a mean (SD) of 2.60 (4.18) doses of ibuprofen over the study period. See eResults 2 in Supplement 2 for more detailed analysis.

Overall, opioid requirements were generally low across both groups. For all study participants, the mean (SD) number of oxycodone doses used cumulatively over the 7-day period was 1.36 (3.05). A total of 8 (8%) participants used more than 5 oxycodone doses, and 4 (4%) (inclusive) used more than 10. No study participant consumed more than 15 oxycodone doses total. Participants in the oxycodone rescue group consumed a mean (SD) of 1.89 (3.80) opioid doses over the 7-day period compared with 0.77 (1.80) in the ibuprofen rescue group (difference, 1.13; 95% CI, –0.11 to 2.36).

Participants in both groups reported similar levels of epistaxis using the VAS scores at all points. A subanalysis comparing ibuprofen users with nonusers also showed no difference in epistaxis severity across all points.

The ongoing national opioid crisis stems not just from the illicit acquisition of narcotics, but also partially from overuse of legal prescriptions. Excessively prescribed opioid medications contribute to prolonged use or abuse by the patient, family members, or friends.^6,7 Consequently, there is increased interest in optimizing the opioid prescribing practices of physicians, including otolaryngologists, to reduce overprescription of narcotics. Endoscopic sinus surgery is one of the most frequently performed surgeries in the US; thus, it is an important contributor to the distribution and procurement of opioids.²⁶

The reported consumption of narcotics after ESS varies widely in the literature, and there is no expected or standardized average of how many doses a patient will require.^9,13,27 As previously mentioned, a recent survey of rhinologists showed that, on average, 27 opioid doses are prescribed after ESS.¹³ Our study found that the actual narcotic requirement for these patients is considerably lower, suggesting a general trend toward overprescribing of narcotics for ESS. While a few patients in our study consumed more than 5 opioid doses total, most patients required no narcotics. Across all patients, the average narcotic requirement was fewer than 2 oxycodone doses throughout the recovery period, which is notably fewer than has been reported in prior studies.^9,13,27 Our data support the efficacy of nonnarcotic analgesics as first-line treatment for pain after ESS, as most patients in this study had sufficient pain control with acetaminophen with or without ibuprofen.

Given the low prevalence of opioid use, it was important to perform a post hoc subanalysis of pain scores in opioid users compared with nonusers. This showed that those who used opioids reported higher pain scores than nonusers across the first 5 PODs. However, we cannot make causative claims, and this finding can be interpreted in several ways. One potential explanation is that in patients who require additional analgesics beyond acetaminophen alone, pain scores are often high enough that they will ultimately require opioids during the postoperative period. The questionnaires asked participants to report their average pain levels over the prior 24 hours, and not just when they were taking an analgesic, so the scores should account for the effect of all medications consumed. The difference in the percentage of acetaminophen-only patients across both groups despite similar pain scores also suggests that the availability of another nonnarcotic (ie, ibuprofen) may lower the threshold and make patients more willing to consume a second-line analgesic. Importantly, opioid use did not sufficiently reduce their average pain scores to comparable levels such as those experienced by participants who did not require opioids. This also highlights the advantage of appropriately used multimodal, nonopioid regimens, which can be more efficacious than narcotics after otolaryngologic procedures.²⁸ The differences observed may illuminate different patient thresholds of pain or degrees of pain that require multimodal therapy. Similar to opioid use, ibuprofen consumption was also associated with higher pain scores.

While some patients in our study successfully used ibuprofen as a second-line therapy, obviating the need for narcotics, our study does not provide definitive evidence that the addition of ibuprofen to an acetaminophen-based pain regimen can reduce opioid consumption, despite a trend toward lower consumption. The possibility of a type 2 error cannot be excluded, based on the unexpectedly high efficacy of acetaminophen as a first-line analgesic therapy and the baseline low rate of narcotic use in the oxycodone rescue group. Despite meeting the goal recruitment numbers for the study based on the power calculation, a larger study may be necessary to adequately power the study and better assess the benefit of narcotic-sparing regimens. However, given the favorable safety and tolerance profile of ibuprofen, it may be reasonable to consider ibuprofen as a second-line analgesic for patients who are intolerant of opioids. Importantly, ibuprofen was not associated with an increased risk of postoperative epistaxis in this study. However, the study was likely underpowered to detect such a difference.

There is a growing body of evidence that nonopioid medications can sufficiently control acute pain following ESS. One study directly compared the efficacy of rofecoxib and hydrocodone-acetaminophen postoperatively and found no difference in pain scores between groups.²⁷ Another study found improved pain control with pregabalin compared with acetaminophen and did not prescribe opioids to either group of patients.²⁹ A double-blinded randomized clinical trial comparing the ability of intravenous ketorolac and fentanyl with control pain when administered during the immediate postoperative period following ESS found no difference in pain control between the 2 treatment groups.³⁰ Notably, despite a theoretical greater concern for epistaxis following the use of NSAIDs postoperatively, no increased incidence of hemorrhage or anemia was found in patients who were randomized to ketorolac. Similarly, our study found no difference in epistaxis severity between ibuprofen users and nonusers. This is despite participants using 600 mg of ibuprofen per dose, which in some cases may be higher than the typical dose when used as a stand-alone analgesic. Further, patients who undergo other procedures performed by otolaryngologists who are high risk for postoperative hemorrhage, such as tonsillectomy, are frequently prescribed ibuprofen without adverse consequences.²⁶

The knowledge gained from this study can be taken in several future directions. Randomized clinical trials that evaluate the synergistic effects of multimodal regimens and compare scheduled nonnarcotic regimens (eg, scheduled acetaminophen) with current regimens can help elucidate whether scheduling medications can further reduce narcotic use.^14,31 Additionally, determining which patients are higher risk for long-term opioid use can help clinicians optimize regimens for each patient. Higher-risk patients may benefit from multidisciplinary specialty care input, scheduled regimens, and potentially other medications, such as tramadol or gabapentin, although these are not without their own adverse effects. Finally, increased opioid education at the time of prescription is imperative to increase patient awareness and promote healthy practices. Increased physician education in this arena is also warranted.³²

There are several limitations to this study. Patients’ willingness to participate in an opioid-related study could itself serve as a source of selection bias. For pragmatic reasons, this study did not require adherence to strict schedules of medication dosing or standard postoperative nonanalgesic medication regimens. In addition, the low levels of narcotic consumption observed may have precluded achieving significance in pain score comparisons across groups, as the pre hoc power analysis predicted higher levels of opioid use. It is possible that the act of enrolling a patient into an opioid study may have influenced participants to reduce opioid use. However, we would have predicted a similar effect on both randomized groups, and thus would not have expected this bias to affect the comparison between groups. Furthermore, the lack of blinding and no placebo may have also introduced conscious or unconscious bias in the patients in assessing their pain levels. However, given the risks and health concerns associated with blinding patients to potential narcotic use, blinding was avoided. Similarly, a placebo was avoided, as introducing a placebo may have led to insufficient pain control. Given the heterogeneity of patient population regarding nasal polyps and differing practice patterns among clinicians, which was out of the scope of this study, steroid administration was left up to surgeon discretion. The potential analgesic effects of steroids, and the lack of assessment for compliance with saline rinses, may have increased the risk for a type 2 error.³³ The loss to follow-up in this study may have introduced selection bias and must be considered when interpreting the results.

Based on the findings in this report in the context of the existing literature and practice standards, it is evident that there are opportunities to modify prescription patterns after ESS. Multimodal analgesic therapy appears effective in managing pain after sinus surgery in most patients, and postoperative pain/discomfort in most patients who undergo ESS can be managed with acetaminophen with or without ibuprofen. The overall narcotic requirement following ESS in this study was low. Although our study did not demonstrate that ibuprofen could reduce narcotic use when serving as a second-line analgesic, any and all potential nonopioid analgesics that are safe and can be used during the postoperative period deserve continued study to reduce the quantity of opioids that are prescribed to, and consumed by, patients who are undergoing ESS.

Accepted for Publication: June 21, 2021.

Published Online: August 5, 2021. doi:10.1001/jamaoto.2021.1839

Corresponding Author: Peter H. Hwang, MD, Division of Rhinology and Endoscopic Skull Base Surgery, Department of Otolaryngology–Head and Neck Surgery, Stanford University School of Medicine, 801 Welch Rd, 2nd Floor, Stanford, CA 94305 (hwangph@stanford.edu).

Author Contributions: Dr Ayoub had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Ayoub, Choby, Turner, Chandra, Shah, Nayak, Patel, Hwang.

Acquisition, analysis, or interpretation of data: Ayoub, Choby, Turner, Abuzeid, Raviv, Thamboo, Ma, Chowdhury, Stokken, O’Brien, Akbar, Shah, Roozdar, Nayak, Patel, Hwang.

Drafting of the manuscript: Ayoub, Choby, Thamboo, Shah, Roozdar, Hwang.

Critical revision of the manuscript for important intellectual content: Ayoub, Choby, Turner, Abuzeid, Raviv, Thamboo, Ma, Chandra, Chowdhury, Stokken, O’Brien, Akbar, Nayak, Patel, Hwang.

Statistical analysis: Ayoub, Ma, Hwang.

Administrative, technical, or material support: Ayoub, Choby, Turner, Abuzeid, Raviv, O’Brien, Akbar, Shah, Nayak, Hwang.

Supervision: Choby, Thamboo, Chandra, Chowdhury, Akbar, Nayak, Patel, Hwang.

Conflict of Interest Disclosures: Dr Choby reported personal fees from Tissium and Intersect ENT outside the submitted work. Dr Akbar reported personal fees from Medtronic and Optinose and grants from Astra Zeneca outside the submitted work. Dr Nayak reported consulting fees from Medtronic, Olympus America, and Cook Medical outside the submitted work. Dr Patel reported personal fees from Medtronic, Optinose, and Regeneration Sanofi outside the submitted work. Dr Hwang reported personal fees from Lyra Therapeutics and Sanofi Regeneron outside the submitted work. No other disclosures were reported.

Data Sharing Statement: See Supplement 3.

Additional Contributions: We thank Noor Ibrahim, MD (Division of Rhinology and Endoscopic Skull Base Surgery, Department of Otolaryngology–Head and Neck Surgery, Stanford University School of Medicine, Stanford, California), Rahul Alapati, BS (Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, Pennsylvania), Laura Sampson (Division of Otolaryngology, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada), and Athenea Pascual (Division of Otolaryngology, Department of Surgery, University of British Columbia, Vancouver, British Columbia, Canada) for their contributions to this study. These individuals did not receive compensation for contributing to this study.