A Comparative Health Utility Value Analysis of Outcomes for Patients Following Septorhinoplasty With Previous Nasal Surgery

Key Points

Question

Do patients who undergo primary septorhinoplasty and revision septorhinoplasty have similar improvements in health utility values (HUVs) following surgery?

Findings

In this cohort analysis that included 185 patients who underwent prior nasal surgery (RNS) before functional rhinoplasty and 278 patients who underwent primary septorhinoplasty (PSRP), HUVs were lower for the RNS cohort throughout the first year postoperatively; however, after 6 months, the 2 cohorts were not significantly different on statistical analysis.

Meaning

Euroqol 5-dimension questionnaire-derived HUVs are a valid predictor of outcomes in all patients who undergo septorhinoplasty and, although outcomes are still significantly improved for both primary and revision septorhinoplasty cohorts, previous surgical history should be considered when performing cost utility analysis on the benefits of functional rhinoplasty.

This cohort study examines whether patients who underwent primary septorhinoplasty and revision septorhinoplasty had similar improvements in health utility values following surgery.

Abstract

Importance

Using health utility values (HUVs) as an outcome measure for surgical procedures, including rhinoplasty, allows for a standardized measurement of the qualitative and financial benefit of that procedure on overall health.

Objective

To use Euroqol 5-Dimension (EQ-5D)-derived HUVs to evaluate nasal obstruction outcomes following surgery between primary rhinoplasty (PSRP) and rhinoplasty with previous nasal surgery (RNS) in a prospective cohort study.

Design, Setting, and Participants

A single-institution prospective cohort study between 2013 and 2017 was carried out including 185 patients who underwent RNS before functional rhinoplasty and 278 patients who underwent PSRP for nasal obstruction. The analysis was carried out between December 2018 and February 2019.

Main Outcomes and Measures

Patients in the RNS cohort and PSRP cohort who completed the EQ-5D questionnaire immediately prior to surgery and postoperatively at 2, 4, 6, and 12 months were included in the study. The EQ-5D scores were converted to HUV scores and compared to evaluate for improvement in health status following surgery.

Results

Of the 185 patients in the RNS cohort, 97 (52%) were women and the mean (SD) age was 42.6 (13.4) years. Of the 278 in the PSRP cohort, 156 (55%) were women and the mean (SD) age was 36.0 (15.8) years. Both cohorts experienced significant improvement in HUV postoperatively. The mean (SD) HUV values were decreased in the RNS cohort versus the PSRP cohort both preoperatively (0.85 [0.16] versus 0.89 [0.14]) and postoperatively at 1 year (0.92 [0.14] versus 0.94 [0.11]); however, the difference in HUV was not statistically significant after 6 months postoperatively. Patients in the RNS cohort were typically older (76 [41%] older than 40 years vs 84 [31%], P = .02) and more likely to have sleep apnea (31 [19%] vs 24 [10%], P = .01) than in the PSRP cohort. On multivariable analysis, the use of spreader grafts and alar rim grafts in the PSRP cohort was significantly associated with improved HUV at 12 months. (r = −0.06; 95% CI, −0.11 to −0.01, P = .03).

Conclusions and Relevance

Patients with a history of prior nasal surgery may represent a unique cohort when assessing health utility outcomes. Euroqol 5-dimension questionnaire-derived HUVs are a valid predictor of outcomes in all patients who undergo septorhinoplasty. Although outcomes were significantly improved for both cohorts, previous surgical history should be considered when performing cost utility analysis on the benefits of functional rhinoplasty.

Level of Evidence

3.

Introduction

Nasal congestion is a prevalent complaint in the general population, affecting more than 10% of the population in the United States and an approximately equal percentage globally.¹ The considerable population burdened with this disease results in costs to the United States health care system estimated at over 3.5 billion dollars annually with considerable impact on quality of life, work productivity, and academic performance.^2,3 Nasal obstruction, a subset of nasal congestion, is a specific complaint that otolaryngologists frequently encounter, particularly when evaluating patients for surgical intervention.

Thorough evaluation and physical examination can identify the exact location of nasal obstruction and determine the appropriate intervention. Pathologic findings at any of these subsites can lead to a different recommendation for treatment.⁴ Appropriate diagnoses can therefore prevent unnecessary surgical interventions. Septoplasty, for example, may be inappropriately recommended to a patient who should undergo a rhinoplasty, making subsequent rhinoplasty more difficult, increasing the need for a second site to obtain cartilage for structural reconstruction, and negatively impacting a patient psychologically.⁵ There are a considerable number of patients who are not satisfied following surgical intervention for nasal obstruction, highlighting the importance of maximizing appropriate diagnosis and treatment.⁶

Health Utility Values (HUVs) are a measure of overall health that is derived from patient valuation of their own health. Health Utility Values have been used as a measure of outcomes following a number of chronic conditions.^7,8,9,10 It allows for the standardized quantification of the health care burden of a specific disease and allows for health policy to determine the practical value of surgical interventions.^7,11 Using HUVs as a measure of outcomes allows for the quantification of detrimental prior failed surgeries and the effect that these failed surgeries have on the valuation of health after subsequent interventions. In this study, patients undergoing primary functional septorhinoplasty (PSRP) as an initial surgical treatment were compared with patients who had previous nasal surgery and subsequently underwent revision rhinoplasty (RNS) to attempt to correct their nasal obstruction. Outcomes were measured using EQ-5D–derived HUVs, as self-reported patient valuation of their overall health before and after surgery.

Methods

The study population consisted of 185 patients who underwent RNS and 278 patients who underwent PSRP without prior surgery for nasal obstruction at a single institution between 2013 and 2017. Of the 185 patients in the RNS cohort, 97 (52%) were women and the mean (SD) age was 42.6 (13.4) years. Of the 278 in the PSRP cohort, 156 (55%) were women and the mean (SD) age was 36.0 (15.8) years. Patients were required to have completed a preoperative, or baseline, EQ-5D questionnaire as well as at least 1 postoperative EQ-5D score. Postoperatively, the same questionnaires were administered to patients at 2, 4, 6, and 12 months. Preoperative and postoperative scores were compared to evaluate for improvement in overall health in both cohorts. This study received local institutional review board approval from the Massachusetts Eye and Ear Human Subjects Research Committee, and informed written consent was obtained from all participants included in the study.

Euroqol 5-Dimension Questionnaires

The EQ-5D questionnaires were used to evaluate preoperative and postoperative overall health scores at the previously specified time points to determine response to surgery. The EQ-5D questionnaire measures health across 5 domains: mobility, self-care, usual activity, pain/discomfort, and anxiety/depression. Each domain was rated on a range from 1 (no problems) to 3 (extreme problems or inability to function). In addition, patients were asked to rate their overall health on a visual analog scale from 1 to 100, with 100 representing best imaginable health and 0 representing worst imaginable health.¹² The EQ-5D scores were used to derive HUVs from population-based data. The EQ-5D scores can be differentiated in 3125 unique health states. These scores can be converted to HUV values using population-based data from national surveys of the adult US population, where each health state has a corresponding HUV.^8,13,14 The HUVs were used to compare outcomes between the 2 cohorts to evaluate the effect of previous nasal surgery.

Statistical Analysis

All statistical analysis was performed using STATA statistical software (version 12.0, STATACorp). χ² tests were used for normally distributed categorical variables, and student t tests were used for continuous variables that were stratified into 2 groups. Analysis of variance (ANOVA) tests were used to compare parametric continuous variables. A Holm-Sidak post hoc pairwise multiple comparison analysis was used to examine whether postoperative HUVs were significantly improved from baseline. For multivariable regression models, all outcome variables were initially tested with univariate analysis. P values of less than .05 were considered statistically significant.

Results

Overall, 185 patients in the RNS cohort and 278 in the PSRP were included in the study population. Compared with patients who had no previous nasal surgery, the cohort of patients who had a history of previous nasal surgery were significantly older (76 [41%] older than 40 years vs 84 [31%] older than 40 years, respectively; P = .02), were more likely to have concurrent sleep apnea (31 [19%] with sleep apnea vs 24 [10%] without sleep apnea, P = .01), and had a lower baseline HUV (mean [SD], 0.85 [0.16] vs 0.89 [0.14], P = .01) (Table 1).

Table 1. Univariate Comparison of Patients Undergoing Rhinoplasty With and Without Previous Nasal Surgery.

Variable	No. (%)		P Value
	Revision Nasal Surgery	Primary Septorhinoplasty
Sex			.55
Male	88 (47.5)	122 (44.7)
Female	97 (52.4)	151 (55.3)
Age, y			.02
<40	109 (58.9)	189 (69.2)
>40	76 (41.1)	84 (30.8)
Sinus disease			.07
Yes	48 (26.2)	52 (19.1)
No	135 (73.8)	220 (81.0)
Sleep apnea			.01
Yes	31 (18.6)	24 (9.9)
No	136 (81.4)	219 (90.1)
Allergies			.78
Yes	102 (56.0)	156 (57.4)
No	80 (44.0)	116 (42.7)
History of nasal fracture			.83
Yes	93 (51.4)	137 (50.4)
No	88 (48.6)	135 (49.6)
Baseline HUV, mean (SD)	0.85 (0.16)	0.89 (0.14)	.01

Abbreviation: HUV, health utility value.

The mean (SD) baseline HUV for the RNS cohort was 0.85 (0.16) compared with 0.89 (0.14) in the PSRP cohort (P = .01). The mean HUVs in the RNS cohort were decreased preoperatively and postoperatively for the entire duration of the study period compared with the PSRP cohort; however, the difference in HUV was not statistically significant after 6 months postoperatively, implying that the RNS cohort improved to a similar level as the PSRP cohort postoperatively. Following surgical intervention, the mean (SD) 2-month HUV was 0.89 (0.15) in the RNS cohort vs 0.92 (0.12) in the PSRP cohort (P = .05), and the mean (SD) 1-year postoperative HUV was 0.92 (0.14) in the RNS cohort vs 0.94 (0.11) in the PSRP cohort (P = .28) (Figure 1).

Figure 1. Mean Health Utility Value at Baseline and Postoperatively Stratified by History of Previous Nasal Surgery.

The change in HUV between the preoperative baseline value and the HUV at 1 year was compared with the change in visual analog scale EQ-5D score to evaluate the correlation between the 2 cohorts (Figure 2). A positive correlation was observed between the change in the 2 variables using Spearman rank order testing (0.23 in the PSRP cohort vs 0.18 in the RNS cohort, P < .01). The linear best fit lines were not significantly different between the 2 cohorts, implying approximately equal sensitivity.

Figure 2. Correlation Between Change in Health Utility Value (HUV) and Change in Visual Analog Scale.

The orange line represents a linear trend line between change in HUV and change in visual analog scale. Each dot represents an individual patient’s change in HUV and corresponding change in visual analog scale.

In the PSRP group, all patients underwent septoplasty at the time of the surgery. Out of the 285 patients in the PSRP cohort, 79 had turbinoplasty, 15 had nasal osteotomies, 204 patients had spreader grafts, 41 had alar rim grafts, 33 had a lateral crural strut graft, and 52 had some additional cosmetic procedure at the time of surgery. On multivariable regression analysis, the use of spreader grafts (r = 0.15; 95% CI, 0.07 - 0.23) and alar rim grafts (r = −0.06; 95% CI, −0.11 to −0.01) was significantly associated with a higher HUV at 12 months. Of the 185 in the RNS cohort, 108 patients had a septoplasty, 26 had turbinoplasty, 10 had osteotomies, 125 had spreader grafts, 31 had lateral crural strut grafts, 15 had alar rim grafts, and 29 had some additional cosmetic procedure at the time of surgery. None of these variables were significantly associated with HUV at 12 months on multivariable regression analysis, both with and without the addition of septoplasty to the model (Table 2).

Table 2. Multivariable Regression Models of Surgical Factors Associated With Health Utility Value at 12 Months for Primary and Revision Septorhinoplasty.

Variable	Primary Septorhinoplasty		Revision Septorhinoplasty
	Regression Coefficient (95% CI)	P Value	Regression Coefficient (95% CI)	P Value
Turbinoplasty	0.04 (−0.01 to 0.09)	.12	0.06 (−0.06 to 0.17)	.33
Osteotomies	0.06 (−0.04 to 0.17)	.25	0.10 (−0.06 to 0.26)	.22
Spreader grafting	0.15 (0.07 to 0.23)	.001	−0.001 (−0.16 to 0.16)	.98
Alar rim grafting	−0.06 (−0.11 to −0.01)	.03	0.04 (−0.11 to 0.19)	.51
Lateral crural strut grafting	0.04 (−0.31 to 0.12)	.52	−0.03 (−0.12 to 0.07)	.57
Additional cosmetic procedure	0.05 (0 to 0.11)	.05	−0.03 (−0.14 to 0.09)	.63

Discussion

Findings of this study suggest that septorhinoplasty is an effective surgical intervention for appropriately selected patients. The disparity between the RNS cohort and the PSRP cohort affirms that previous nasal surgery can hamper patient outlook on future surgeries. This finding is in contrast to sinus surgery, where HUVs are not affected by previous surgery.⁹ The finding on multivariable analysis that spreader grafts and alar rim grafts were significantly associated with an improvement in HUV at 12 months postoperatively supports the use of cartilage grafting as high utility in achieving long-term improved structural support. The emotional ramifications of requiring revision rhinoplasty have previously been explored, with patient expectations playing a significant role in outcomes.⁵ When evaluating satisfaction with rhinoplasty specific surveys, 1 study¹⁵ had findings that mirrored the results in this study in that both revision and primary rhinoplasty patients experienced positive outcomes, but that primary rhinoplasty patients had higher levels of satisfaction after their surgery.

Studies showing that primary and revision surgical patients represent unique population when considering outcomes are important for counseling patients and establishing realistic expectations. Health utility values are particularly useful because individuals are able to value their overall health in a standardized fashion to better compare the utility of surgical interventions and allow for the valuation of health care. In an attempt to maximize quality of life for these patients, accurate diagnosis is important in initially evaluating nasal obstruction. The differential diagnoses of nasal obstruction are extensive, and include mechanical, infectious, and allergic conditions.¹⁶ Appropriate diagnosis is crucial because this dictates medical vs surgical intervention, and can be difficult if the clinician is not familiar with the full spectrum of diagnostic techniques for thoroughly evaluating nasal obstruction.¹⁷ This study further demonstrates the importance of appropriate diagnosis because improper surgical intervention can be detrimental to both short- and long-term quality of life. Inappropriate diagnosis of obstruction that is attributed to a deviated septum, for example, can lead to inappropriate septoplasty, decreased baseline HUVs at the time of presentation for rhinoplasty, and affect long-term benefit in HUVs following subsequent rhinoplasty. It is notable, however, that the cohorts were not significantly different at the 2 year mark, implying that, over time, it may be possible to achieve the same health state. In particular, those who underwent previous nasal surgery and subsequent rhinoplasty improved to a greater degree from their depressed baseline HUV level to approximately equal health state compared with the cohort that had no previous surgery.

Using HUV as an outcome measure provides valuable insight into the overall detrimental impact of a disease process for a patient because that patient is able to quantify the overall negative influence of living with a disease or condition. In addition, the patient effectively values their health in a standardized fashion that allows for comparisons with distinct disease entities.^9,11 Health utility values reflect an individual’s health preferences when presented with different health outcomes. These utility values range from 0 to 1, with 1 reflecting perfect health. Once defined for a particular condition, these values can be combined with survival estimates and quality-adjusted life-years (QALYs) for use in cost-utility analyses of health care interventions. In health economics, the “standard gamble” approach allows individuals to decide between a certain health state (for example, chronic nasal obstruction) or gamble for 1 better (for example, full health) vs 1 that is worse (for example, worsened nasal obstruction or a new health state with a complication from surgery).¹⁸ A standardized system is of particular value when attempting to quantify the burden of disease. Quality-adjusted life years are the recognized standardized unit to evaluate for benefit of health care interventions.¹⁹ The HUVs provide a means to measure the quality improvement in health for a particular intervention. This, however, is complicated by the fact that an individual’s valuation of their own health depends on a number of variables including fear of the unknown, previous attempts at intervention, psychological state, past experience, and cultural backgrounds.¹¹ As a result of this study, HUVs for revision septorhinoplasty have been established. With established HUVs, future studies can focus on using HUVs to establish QALYs for cost-utility analyses. These can be used to further understand, from a health economics perspective, when it is appropriate and cost-effective to treat a patient medically vs surgically for nasal obstruction.

Limitations

This study has several limitations. Patients included in this study completed preoperative and postoperative EQ-5D surveys. Those who were lost to follow-up may have had unsatisfactory outcomes. The time between previous surgery and presentation for the RNS cohort was not collected as part of the study. The length of time between the patient receiving previous surgery and presenting for a revision surgery may influence baseline and postoperative HUVs. Because this study highlighted the impact of previous failed therapies, these patients had variable preoperative medical therapies to treat nasal obstruction that were not recorded as a part of this study, which may have altered their HUV scores. Despite these limitations, this study incorporated 2 relatively large cohorts of patients undergoing septorhinoplasty and provides insight into the relative improvement of HUV for each of these cohorts.

Conclusions

Patients who undergo prior nasal surgery represent a unique cohort when assessing outcomes following rhinoplasty. The EQ-5D–derived HUVs are valid predictors of patient outcomes in patients regardless of their history of prior nasal surgery. Although this study shows that outcomes are significantly improved for both patients with and without prior history of nasal surgery who undergo rhinoplasty, prior surgical intervention should be considered when performing cost-utility analyses on the benefits of functional rhinoplasty. This study highlights the importance of appropriate diagnosis for patients presenting with nasal obstruction.