Trends and complications in functional endoscopic sinus surgery and balloon sinuplasty: A TriNetX database analysis

Abstract

Objectives

Balloon sinuplasty (BSP) and functional endoscopic sinus surgery (FESS) have undergone changes in utilization over time. This study investigates national trends in BSP and FESS and postoperative outcomes over the last decade as well as the impact of Corona Virus Disease 2019 (COVID‐19) on these trends.

Methods

The TriNetX database was queried for patients undergoing either FESS or BSP from 1/1/2011 to 5/4/2024. Data were collected on patient demographics, diagnoses, and postoperative outcomes.

Results

A total of 1738 patients underwent BSP and 90,311 underwent FESS. A greater proportion of FESS patients had diabetes (p = 0.001), hypertension (p < 0.001), and chronic pulmonary disease, p = 0.001) compared to BSP whereas a similar proportion of patients for both cohorts had a history of ischemic heart disease (p = 0.73). Body mass index was higher for patients undergoing FESS (28.6 ± 6.5) compared to BSP (28.1 ± 6.578; p = 0.02). Overall postoperative complications were similar between BSP and FESS (odds ratio [OR] = 0.80, 95% confidence interval [CI] = 0.58–1.09) with similar rates of epistaxis (OR = 0.77, 95% CI = 0.53–1.12) and cerebrospinal fluid (CSF) leak (OR = 0.14, 95% CI = 0.01–2.25). Overall revision rate was higher for FESS (OR = 0.34, 95% CI = 0.21–0.57). Thirty‐day readmission (OR = 0.44, 95% CI = 0.29–0.66) and Emergency Department visits (OR = 0.51, 95% CI = 0.28–0.92) were less common in patients who received BSP. Surgical volume consistently increased over time for both cohorts but at a more rapid pace for BSP (425.64%) compared to FESS (274.19%) and a dramatic decrease in volume by 44.85% and 22.28%, respectively, at the onset of COVID‐19.

Conclusion

Overall, BSP and FESS surgical volume have steadily increased over time with a drastic reduction following the COVID‐19 pandemic. BSP and FESS carry different complication profiles emphasizing the importance of patient selection and preoperative counseling.

Level of Evidence

Level 4.

INTRODUCTION

Functional endoscopic sinus surgery (FESS) and balloon sinuplasty (BSP) are the two mainstay minimally invasive surgical treatment options for the management of chronic rhinosinusitis (CRS). Before considering surgical intervention, however, patients typically trial medical therapy, with persistent symptoms. Traditionally, FESS has been the gold standard surgical option for the treatment of CRS until BSP was approved by the US Food and Drug Administration (FDA) in 2005 with a formal introduction of BSP‐specific Current Procedural Terminology (CPT) codes in 2011. ¹ , ² Although similar to FESS, indications for BSP specifically include radiographic and exam findings consistent with CRS without nasal polyp (CRSsNP) refractory to medical therapy. ³ , ⁴ Since its introduction, there has been an increase in BSP with surgeons performing the procedure in the operating room as well as in the office. ¹ , ⁵ , ⁶ Increasing utility of BSP is multifactorial with some arguing comparable efficacy for treatment of CRS relative to FESS given avoidance of mucosa and bone removal and subsequent formation of postoperative synechiae. ⁷ , ⁸ , ⁹ It may also be considered for patients with multiple comorbidities in which general anesthesia is determined to be unsafe for the patient. Some suspect economic reasons pointing to reimbursement practices for office‐based procedures. ¹⁰

Despite increasing popularity of BSP, FESS remains the most performed procedure for CRS. Complications related to FESS are well documented including epistaxis, vision changes due to orbital injury, and skull base injury with subsequent cerebrospinal fluid (CSF) leak. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ However, there is a paucity of data evaluating complications related to BSP as well as comparing rates of complications to FESS. ² , ⁶ , ¹⁶ , ¹⁷ , ¹⁸ Reported complications for BSP also include CSF leak, orbital complications, and severe bleeding. ⁶ Although trends in BSP compared to FESS have previously been evaluated, the impact of Corona Virus Disease 2019 (COVID‐19) on these trends has yet to be discussed. ¹⁰ The current study expands on limited previous database studies by describing current trends in surgical treatment of CRS through 2023, including geographic distribution, postoperative complications, and rates of revision as well as the impact of COVID‐19 on these trends.

METHODS

The TriNetX database is an electronic healthcare record database representing over 250 million patients across 120 healthcare organizations (HCOs) from 19 different countries in North and South America, Europe, Africa, the Middle East, and Asia‐Pacific. This database contains deidentified patient information which is extracted from participating HCO electronic medical records and is updated multiple times each month. Additionally, the TriNetX database maps correlated International Classification of Diseases Ninth Revision, Clinical Modification (ICD‐9‐CM), and ICD‐10‐CM codes, seamlessly combining both codes into the same query. ¹⁹ This mapping has been reviewed for clinical accuracy and validity by licensed physicians. ¹⁹ The database was queried for patients undergoing either FESS (CPT 31253, 31254, 31255, 31256, 31257, 31259, 31267, 31276, 31287, 31288) or BSP (CPT 31295, 31296, 31297) since 1/1/2011 until 5/4/2024 (Supporting Information S1: Supplement I). Concurrent BSP and FESS were excluded from the analysis.

Data were collected on patient demographics and postoperative outcomes globally. Outcomes were chosen a priori based on previous studies and included 90‐day postoperative complications such as CSF (ICD‐10‐CM G96.0), orbital injury and sequelae (ICD‐10‐CM H49‐H54, H00.039, H05.0, H04.2, H05.239, H11.33, and S05.10XA), epistaxis (ICD‐10‐CM R04.0), meningitis (ICD‐10‐CM G00.3, G00.2, G00.9, G04, G04.2, and G03) and reoperation, as well as 30‐day readmission, and return to the Emergency Department (ED) after discharge. ¹ , ² , ¹⁷ , ²⁰ Only patients with a new diagnosis of each outcome or complication were included in the analysis. Statistical analysis was performed using integrated software within the TriNetX software using measures of association including chi‐square tests and odds ratios (ORs) with 95% confidence intervals (CIs). Unadjusted OR was calculated with the Haldane–Anscombe correction to compensate for outcomes with incidence of 0. Statistical significance was determined to be p < 0.05.

RESULTS

A total of 1738 patients underwent BSP and 90,311 underwent FESS (Table 1), of which 93% total were treated within the United States. BSP demonstrated a female predominance of 51% compared to 48% (p = 0.01). Patients undergoing BSP were younger (40.8 ± 23.8) compared to FESS patients (47.2 ± 19.0; p < 0.001). A greater proportion of patients were white in the BSP cohort (77%) compared to the FESS group (70%) (p < 0.001). The proportion of Asian patients was similar between the cohorts (p = 0.32). A smaller proportion of patients was Black/African American in the BSP group (7%) compared to the FESS cohort (10%) (p < 0.001). A greater proportion of patients with pre‐existing comorbidities including diabetes (11% vs. 8%, p = 0.001), hypertension (27% vs. 23%, p < 0.001), and chronic pulmonary disease (26% vs. 24%, p = 0.001) underwent FESS compared to BSP, whereas a similar proportion of patients undergoing BSP and FESS had a history of ischemic heart disease (7%, p = 0.73). Body mass index (BMI) was higher for patients undergoing FESS (28.6 ± 6.5) compared to BSP (28.1 ± 6.8) (p = 0.02). A greater proportion of patients who underwent BSP had a history of acute rhinosinusitis (30% vs. 25%, p < 0.001). Equal proportions of patients undergoing either BSP or FESS had a history of CRS (76% vs. 75%, p = 0.47); however, a greater proportion of patients undergoing FESS had CRS with nasal polyposis (CRSwNP) compared to BSP (26% vs. 6%, p < 0.001). During the study period, BSP was predominately performed in the South (51%), followed by the Northeast (22%), West (15%), and finally, the Midwest (9%). FESS was performed most often in the South (40%), closely followed by the Northeast (29%) with lower rates in the West (16%) and Midwest (13%).

Table 1.

The demographics of patients undergoing BSP or FESS.

Demographics	BSP (n = 1738)	FESS (n = 90,311)	p value
Age (years, mean ± SD)	40.8 ± 23.8	47.2 ± 19.0	<0.001
Sex
Female	883 (51%)	42,992 (48%)	0.008
Male	847 (49%)	44,550 (49%)	0.62
BMI	28.1 ± 6.8	28.6 ± 6.5	0.02
Ethnicity
Non‐Hispanic	1285 (74%)	64,517 (71%)	0.02
Hispanic	186 (11%)	7289 (8%)	<0.001
Race
White	1339 (77%)	63,640 (70%)	<0.001
Black	130 (7%)	8880 (10%)	0.001
Asian	51 (3%)	3042 (3%)	0.32
Other	173 (10%)	10,332 (11%)	0.053
Comorbidity
Diabetes	143 (8%)	9649 (11%)	0.001
Ischemic heart diseases	118 (7%)	6323 (7%)	0.73
Chronic lung diseases	411 (24%)	23,848 (26%)	0.001
Hypertension	394 (23%)	24,082 (27%)	<0.001
Geographic distribution
Northeast	383 (22%)	25,798 (29%)
Midwest	161 (9%)	11,895 (13%)
South	889 (51%)	36,446 (40%)
West	256 (15%)	14,411 (16%)

Abbreviations: BMI, body mass index; BSP, balloon sinuplasty; FESS, functional endoscopic sinus surgery; SD, standard deviation.

Overall postoperative complications (Table 2) were similar between BSP (2.89%) and FESS (3.6%) (OR = 0.80, 95% CI = 0.58–1.09) with similar rates of epistaxis (OR = 0.77, 95% CI = 0.53–1.12) and CSF leak (OR = 0.14, 95% CI = 0.01–2.25) on subanalysis. Need for additional surgical interventions including orbital surgery (OR = 0.43, 95% CI = 0.03–6.89), control of epistaxis (OR = 0.63, 95% CI = 0.36–1.12), and skull base repair/lumbar drain placement (OR = 0.14, 95% CI = 0.01–2.29) were also similar between the two cohorts. There was no difference in the use of biologic therapies postoperatively between the two cohorts. Patients who underwent BSP were more likely to develop meningitis (OR = 2.98, 95% CI = 1.57–5.64) but less likely to sustain an orbital injury intraoperatively (OR = 0.56, 95% CI = 0.32–0.97). Overall revision rate was higher for FESS (2.42%) compared to BSP (0.85%) regardless of revision surgery type (OR = 0.34, 95% CI = 0.21–0.57). Revision BSP was more common in patients previously treated with BSP (OR = 1518.8, 95% CI = 90.56–25,471.76); however, revision FESS was less common in patients previously treated with BSP (OR = 0.01, 95% CI = 0.001–0.17). Thirty‐day readmission (OR = 0.44, 95% CI = 0.29–0.66) and ED visits (OR = 0.51, 95% CI = 0.28–0.92) were less common in patients who received BSP compared to FESS.

Table 2.

The postoperative outcomes of patients undergoing BSP or FESS.

Outcome	BSP (n = 1738)	FESS (n = 90,311)	OR (95% CI)	p value
Postoperative complications (90 days)	40 (2.89%)	2465 (3.6%)	0.80 (0.58–1.09)	0.17
Epistaxis	28 (1.69%)	1869 (2.18%)	0.77 (0.53–1.12)	0.18
Cerebrospinal fluid leaka	0 (0%)	182 (0.21%)	0.14 (0.01–2.25)	0.16
Meningitis	10 (0.58%)	174 (0.20%)	2.98 (1.57–5.64)	0.001
Orbital injury	13 (0.86%)	1170 (1.54%)	0.56 (0.32–0.97)	0.04
Surgical interventions following complication (90 days)
Orbital surgerya	0 (0%)	60 (0.07%)	0.43 (0.03–6.89)	0.55
Control of epistaxis	12 (0.69%)	979 (1.08%)	0.63 (0.36–1.12)	0.12
Skull base repair/lumbar draina	0 (0%)	181 (0.2%)	0.14 (0.01–2.29)	0.17
Revision surgery (90 days)	15 (0.85%)	2368 (2.42%)	0.34 (0.21–0.57)	<0.001
Revision BSPa	14 (0.81%)	0 (0%)	1518.8 (90.56–25,471.76)	<0.001
Revision FESSa	0 (0%)	2328 (2.58%)	0.01 (0.001–0.17)	0.001
Biologic therapy initiation (90 days)a	0 (0%)	264 (0.3%)	0.10 (0.01–1.56)	0.10
Readmission (30 days)	23 (1.32%)	2696 (2.99%)	0.44 (0.29–0.66)	0.001
Emergency room visits (30 days)	11 (0.94%)	1197 (1.83%)	0.51 (0.28–0.92)	0.03

Abbreviations: BSP, balloon sinuplasty; CI, confidence interval; FESS, functional endoscopic sinus surgery; OR, odds ratio.

^aHaldane–Anscombe correction.

Surgical volume consistently increased overtime for BSP and at a more rapid pace compared to FESS, with a peak increase in 2012 at 74.36% (Figure 1). Surgical volume for FESS also consistently increased overtime, albeit at a slower rate with a peak increase of 28.68% in 2015. Both surgical volumes for BSP and FESS suffered a dramatic decrease at the onset of COVID‐19 in 2020 with a sharp decline of 44.85% for BSP and 22.28% for FESS. Recovery in surgical volume has stagnated for sinus procedures without a return to pre‐pandemic number of surgical cases.

Figure 1.

Year over year percent change in balloon sinuplasty (BSP) and functional endoscopic sinus surgery (FESS).

DISCUSSION

This study analyzed national trends in FESS and BSP patient demographics, utilization, as well as postoperative outcomes since the introduction of BSP CPT coding in 2011. Our analysis expands on previous population‐based studies examining trends in BSP and in the US and provides the most up‐to‐date information from the years 2011 to 2024. This study period is both meaningful and strategically chosen because it represents the longest retrospective study comparing trends between BSP and FESS allowing for comprehensive side‐by‐side analysis of the growing use of these two procedures. Additionally, the study period includes the post‐COVID‐19 era, which has yet to be evaluated, allowing for analysis of the impact of COVID‐19 on these two elective procedures.

When evaluating patient demographics, a greater proportion of patients with at least one comorbidity (diabetes, chronic lung disease, and hypertension) underwent FESS compared to BSP. Additionally, BMI was also greater for patients undergoing FESS. FESS is often performed in the operating room, whereas isolated BSP can be done in either the clinic or the operating room. Although in the present study, we were unable to analyze the setting of surgery, previous literature corroborates this notion. Chaaban et al. found that a majority of BSP is performed in the clinic (86.53%) instead of the operating room, whereas FESS was almost exclusively performed in the operating room (99.26%). ¹ We hypothesize a greater proportion of patients undergoing FESS have pre‐existing conditions compared to BSP because these procedures are done in the operating room with the assistance of an anesthesiologist where it is easier to medically manage the patients' conditions. BSP, on the other hand, is more often done in the office where patients are typically healthier patients that do not require monitored anesthetic care. In the present study, patients undergoing BSP were also younger than patients undergoing FESS. This is in line with previous findings by Ference et al. who performed a comparative study between FESS and BSP and observed patients undergoing BSP were also younger. ²¹ We hypothesize BSP patients are younger than patients undergoing FESS secondary to more severe disease with increasing age. A previous study exploring the impact of age on CRS disease showed that older patients (age 60–77) had significantly worse disease on computed tomography (CT). ²² , ²³ This has been attributed to age‐related persistence of impairment of innate immunity with dysfunction of the epithelium. ²³

Complications in FESS have been well documented with the most severe including epistaxis, orbital injury, and skull base injury with resultant CSF leak and/or meningitis. ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ²⁴ Similar complications have also been documented in BSP but to a much more limited extent with even fewer studies comparing rates of complications between BSP and FESS. ² , ⁶ , ¹⁶ , ¹⁷ , ¹⁸ A recent study evaluated similar complication profiles over a decade ago and found overall complication rate were 7.35% for FESS and 5.26% for BSP. ¹ In the present analysis, rates of complication were lower for both FESS (3.6%) and BSP (2.89%) and not statistically different from each other. These rates are likely lower given these rates only included complications reported within the first 90 days after surgery whereas the study by Chaaban et al. included reported complications over a 6‐month period. ¹

On subanalysis, BSP patients were at greater risk of developing meningitis following surgery. This may be due to the creation of a false passage with occult fracture of the skull base that is not easily recognized at time of injury. Previous reports evaluating complications between the two procedures did not compare rates of meningitis, however, rates of skull base injuries (CSF leak and pneumocephalus) and central nervous system complications were combined and found to be relatively similar with rates of 0.35% and 0.39% for BSP and FESS, respectively. ¹ We additionally evaluated postoperative CSF leak between the two procedures and found no difference. In isolation, the risk of CSF leak was also low in our study for both procedures. Orbital injury was found to be more common in FESS compared to BSP. Chaaban et al. also found orbital injuries were more common in FESS compared to BSP with rates of 3.47% and 2.95%, respectively. ¹ FESS clearly requires more dissection around the orbit placing the orbit at greater risk of injury. The risk of injury to the orbit during FESS may also be secondary to the use of powered tools (e.g., microdebrider and powered drill), which are used during FESS and not BSP. Chaaban et al. found epistaxis was also more common in patients who underwent FESS with a rate of 3.46%, whereas epistaxis occurred in 2.03% of BSP cases. ¹ Rate of epistaxis was slightly lower and not significant in the present study with a rate of 2.18% for FESS and 1.69% for BSP. Again, FESS requires more dissection with more potential to damage surrounding vasculature and cause subsequent bleeding.

ED and readmission rates were higher for FESS in our analysis. A previous study evaluating ED visits and inpatient admission following sinonasal surgery in which the primary procedure was a sinus or nasal surgery found that almost 10% of initial ED visits and readmissions and 7% of the second ED visits and readmissions were due to postoperative pain or fever, vomiting, and dehydration. ¹¹ Although we are unable to analyze the reason for ED visits and readmission in the TriNetX database, we suspect the avoidance of tissue removal in BSP improves postoperative recovery including acute pain and contributes to the above findings. ⁸ , ⁹ Further analysis is warranted to confirm the etiology behind ED visits and readmission rates for both procedures.

BSP is often reserved for patients with limited sinus disease. This is made apparent by a study by Cooper et al. who found patients with higher Lund‐Mackay scores, radiographic evidence of neo‐osteogenesis, prior sinus surgery, CRSwNP, and gram negative infections were at higher risk of revision surgery after BSP. ²⁵ Our study found overall revision surgery was higher for FESS with 2.42% of cases requiring surgery within 3 months compared to a rate of 0.85% in the BSP cohort regardless of revision surgery type. These rates are lower than the rates reported in the literature. When independently evaluated, FESS revision rates in the literature are variable, ranging from 1% to 15% whereas BSP revision rates have been reported to be around 3.5%. ²⁶ , ²⁷ , ²⁸ , ²⁹ Chaaban et al. compared 6‐month revision rates for BSP and FESS and reported higher rates of 7.89% and 16.85%, respectively. ¹ A consistently higher revision rate for FESS over BSP can partially be attributed to more severe disease at the time of initial surgery requiring tissue removal and subsequently worsening the risk for stenosis and development of adhesions. ³⁰

Since the introduction of BSP CPT codes, there has been a steady increase in BSP procedures performed until the COVID‐19 pandemic, with a peak increase in 2012 at 74.36% with an overall increase during the study period of 425.64%. Although not at the same rate, FESS has also gradually increased, with a peak growth of 28.68% during 2015 and an overall increase of 274.19%. Koester et al. attempted to identify the impetus for the more rapid growth of BSP compared to FESS and found reimbursement for FESS increased by 23% ($3.6–$4.5 million), whereas BSP payments increased by 357% ($9.1–$40.5 million). ³¹ At the onset of the COVID‐19 pandemic, BSP volume decreased by 44.85% and FESS volumes decreased by 22.28% of which both have failed to recover to pre‐pandemic volumes. Prior studies have demonstrated a recovery in most surgical volumes except in some otolaryngology procedures. ³² , ³³ This is likely due to the exposure risk to respiratory pathogens during surgery involving the head and neck. ³⁴ Many institutions implemented restrictions on scheduling surgical cases including delaying nonemergent cases such as endoscopic sinus surgeries. ³⁵ , ³⁶ , ³⁷ To our knowledge, this study is the first to demonstrate fluctuations in FESS and BSP volumes nationally secondary to the COVID‐19 pandemic.

Inherent to most database studies, the present study contains several limitations worth mentioning. The TriNetX database contains deidentified patient information which may result in inaccurate or missing information, inclusion bias, and limited variables. The study only examined 90‐day postoperative complications and subsequent surgical interventions, 90‐day reoperation, 30‐day readmission, and ED visits. We therefore may be underestimating the rate of each outcome. Within the database, there are patients who underwent both FESS and BSP, however, we excluded these patients for both simplicity but also to isolate complications to a specific surgical technique. Often these procedures are performed together, therefore, our sample size underestimates the number of patients undergoing these procedures. The TriNetX database sources patient information only from participating HCOs. This data‐sharing agreement provides institutions with access to the database in return for patient data. Unfortunately, this excludes smaller practices where more limited surgery is performed on healthier patients. There is likely a disproportionate underreporting of BSP procedures and complications compared to FESS. Additionally, patients who did not present to the ED or patients who return to a different HCO that does not participate in TriNetX data sharing may not be captured in our analysis again underestimating outcomes. Finally, objective data specific to CRS severity, including SNOT‐22, Lund Mackay, and Lund Kennedy scores, prevents us from determining the severity of the disease, which would provide more insight into the rates of outcomes between both cohorts.

CONCLUSION

BSP and FESS may both cause epistaxis, meningitis, or orbital injury, as well as require similar rates of revision surgery, emphasizing the importance of informing patients of potential complications during their preoperative counseling. It is also important, when discussing each surgical option with the patient, to educate on the increased risk for Emergency Room visits, readmission, and orbital injury for FESS, whereas BSP has fewer overall postoperative complications. When selecting patients, otolaryngologists should also consider patient comorbidities. Additionally, BSP and FESS surgical volume has steadily increased over time until the COVID‐19 pandemic, which caused a drastic reduction in volume without recovery to date.

AUTHOR CONTRIBUTIONS

Benjamin F. Bitner: Conceptualization; data curation; formal analysis; methodology development; drafting; manuscript preparation. Sina J. Torabi: Conceptualization; formal analysis; methodology development; drafting; manuscript preparation. Theodore V. Nguyen: Formal analysis; methodology development; drafting; manuscript preparation. Jonathan C. Pang: Formal analysis; methodology development; drafting; manuscript preparation. Edward C. Kuan: Methodology development; supervision; manuscript preparation; critical revisions.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest.

ETHICS STATEMENT

The study was exempt from ethics committee review as all data was publicly available.

Supporting information

Supporting information.

Bitner BF, Torabi SJ, Nguyen TV, Pang JC, Kuan EC. Trends and complications in functional endoscopic sinus surgery and balloon sinuplasty: A TriNetX database analysis. World J Otorhinolaryngol Head Neck Surg. 2025;11:368‐374. 10.1002/wjo2.222

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.