Customize your JAMA Network experience by selecting one or more topics from the list below.
Strazdins E, Nie YF, Ramli R, et al. Association of Mental Health Status With Perception of Nasal Function. JAMA Facial Plast Surg. 2017;19(5):369–377. doi:10.1001/jamafacial.2017.0459
Is mental health status associated with the perception of nasal function compared with objective measures?
A cross-sectional study of preoperative patients undergoing rhinoplasty demonstrated that those with poor mental health status had a significantly poorer self-perception of nasal function compared with mentally healthy patients, although the results of nasal airflow analyses were similar.
Poor mental health status is associated with a poorer perception of nasal function; thus, objective assessment of nasal airflow may be advantageous to validate patient reports.
Mental health issues are thought to be overrepresented among patients undergoing rhinoplasty and may be associated with patient presentation prior to surgery.
To assess the association of poor mental health with perception of nasal function.
Design, Setting, and Participants
A cross-sectional study of patients presenting for airway assessment was performed from December 1, 2011, to October 31, 2015, at 2 tertiary rhinoplasty centers in Sydney, Australia. Mental health was independently defined preoperatively by the Mental Component Summary of the 36-item Short Form Health Survey version 2 (a score of <40 indicated poor mental well-being), the Rosenberg Self-Esteem Scale (a score of <15 indicated low self-esteem), and the Dysmorphic Concerns Questionnaire (a score of >11 indicated above-average dysmorphic concerns).
Main Outcomes and Measures
Nasal function was assessed with patient-reported outcome measures, including the Nasal Obstruction Symptom Evaluation Scale, the 22-item Sinonasal Outcome Test, a visual analog scale to rate ease of breathing on the left and right sides, and Likert scales to assess overall function and nasal obstruction. Nasal airflow was assessed by nasal peak inspiratory flow, nasal airway resistance, and minimum cross-sectional area.
Among 495 patients in the study (302 women and 193 men; mean [SD] age, 36.5 [13.6] years), compared with patients with good mental health, those with poor mental health had poorer scores in all patient-reported outcome measures, including the visual analog scale for the left side (mean [SD], 51  vs 42 ; P = .001), visual analog scale for the right side (mean [SD], 54  vs 45 ; P < .001), Nasal Obstruction Symptom Evaluation Scale (mean [SD], 2.64 [0.95] vs 1.96 [1.04]; P < .001), 22-item Sinonasal Outcome Test (mean [SD], 2.14 [0.84] vs 1.33 [0.83]; P < .001), nasal obstruction (58 of 145 [40.2%] vs 83 of 350 [23.7%] with severe or worse obstruction; P < .001), and nasal function (72 of 145 [49.7%] vs 111 of 350 [31.8%] with poor or worse function; P < .001). Subclinical differences in nasal peak inspiratory flow could be demonstrated, but all other nasal airflow measures were similar. Low self-esteem produced a similar pattern, but dysmorphia did not.
Conclusions and Relevance
Poor mental health status is associated with a poorer self-perception of nasal function compared with those who are mentally healthy with clinically similar nasal airflow. Clinicians should be aware that patients with poor mental health reporting obstructed airflow may in part be representing an extension of their negative emotions rather than true obstruction and may require further assessment prior to surgery.
Level of Evidence
Rhinoplasty requires consideration of both functional and cosmetic outcomes. In particular, regarding nasal breathing, surgeons are frequently reliant on patient self-reports to assess these concerns preoperatively. However, patients may report dissatisfaction with their nose owing to many other factors, including cultural ideals, past experiences, and possibly their mental health.
Despite the surgical field gaining an interest in mental health, the literature regarding the psychological aspects of rhinoplasty is conflicting. Many studies over the past 5 decades report that patients undergoing rhinoplasty have poor mental health.1-7 In contrast, other studies view patients undergoing rhinoplasty as psychologically stable,2,8,9 with normal self-esteem but poor body image.3,10-15 A more conservative approach suggests that, as a whole, the population of patients undergoing rhinoplasty have good mental health, but psychological abnormalities, such as body dysmorphic disorder (BDD),4,16,17 anxiety,2 and depression,4,16 are overrepresented. This finding becomes important when considering that impairments in mental health may create cognitive frameworks for organizing and interpreting information, which can influence attention, encoding, interpretation, and recall.18,19 This framework has the potential to affect perceptions by patients of their body and, thus, promote more negative evaluations of their nasal function.
Body dysmorphic disorder is a psychiatric disorder characterized by a preoccupation with an absent or minimal deformity to the extent of interfering with mental, social, and occupational function.4 More important, rhinoplasty is the most common surgical procedure for patients with BDD.4,20-22 Although BDD has an estimated prevalence of 0.5% to 3% of the population,23 patients undergoing cosmetic rhinoplasty have a disproportionate prevalence of the disorder, with estimates ranging from 2% to 15%3,23,24 and more than 30% in some studies.17,25 Furthermore, while not meeting the full diagnostic criteria of BDD, researchers have noted that 40% to 70% of patients undergoing rhinoplasty have greater dysmorphic concerns than seen in the general population.16,17,23,25 Therefore, BDD is an important aspect of mental health for patients undergoing rhinoplasty, being highly prevalent and possessing a spectrum of severity. Although most previous studies have focused on BDD and cosmesis, little is known about how this mental disorder may influence the perception of nasal function.
Mental health is complex and, more importantly, may influence a patient’s perception of his or her nasal function. It is disputed whether the mental health status of patients undergoing rhinoplasty, as a group, is different than that in the general population, and conclusions have often been founded on a questionable evidence base. Most studies indicate that a sizeable proportion of patients undergoing rhinoplasty have mental health disorders. However, to our knowledge, no studies have rigorously considered whether such differences in mental health status affect patient perceptions of nasal function, and none has compared patient-reported outcomes with objective measures in this context. The aim of this study was to assess if mental health status, as determined by mental well-being, self-esteem, and dysmorphic concerns, is associated with baseline perception of nasal function when compared with objective airflow analysis.
Consecutive patients with functional and cosmetic indications for rhinoplasty who underwent nasal airflow analysis from December 1, 2011, to October 31, 2015, at 2 tertiary clinics in Sydney, Australia, were included in a cross-sectional study. This group included patients with breathing difficulties, some of whom were undergoing evaluation, considering surgery, or had previously undergone surgery. Basic demographic information and results of mental health assessments and nasal function questionnaires were collected. Ethics approval was obtained from the St Vincent’s Hospital institutional review board (SVH 09/083). Patients provided verbal consent.
Three validated questionnaires independently defined mental health status. The 36-item Short Form Health Survey version 2 assessed quality of life through physical, social, psychological, and emotional domains,26 with the subset of questions assessing emotional quality of life and well-being forming the Mental Component Summary (MCS). A score of less than 40 on the MCS defined poor mental well-being. The Rosenberg Self-Esteem Scale (RSES) had 10 items for which patients used a 4-point Likert scale from 0 (strongly disagree) to 3 (strongly agree) to answer questions related to self-esteem.27 The RSES is the most widespread measure of self-esteem in research,27 validated in plastic surgery28 and cosmetic surgery29 and the psychological setting,27 but application to patients undergoing rhinoplasty has been limited.15 A score on the RSES of less than 15 defined low self-esteem. The Dysmorphic Concerns Questionnaire is a screening tool using 7 items in which patients rated their concern about their physical appearance relative to others using a 4-point Likert scale from 0 (much less than other people) to 3 (much more than other people).16,30 A score of more than 11 was indicative of above-average dysmorphic concerns.
Five tools assessed patient perceptions of nasal function. A visual analog scale (VAS) asked patients to rate ease of breathing on a linear scale ranging from no nasal obstruction (0 mm) to complete obstruction (100 mm) on the left and right sides of the nose. A number from 0 to 100 was then obtained for severity of nasal obstruction. Two validated questionnaires, the Nasal Obstruction Symptom Evaluation Scale (NOSE)31,32 and the 22-item Sinonasal Outcome Test (SNOT22),31,33 were also used. In addition, the results of a 6-point Likert scale for nasal obstruction ranging from 0 (no problem) to 5 (very severe problem) were recorded. A global score of nasal function anchor scores used a 13-point Likert scale from −6 (terrible) to 0 (neither good nor bad) to 6 (excellent) to assess overall nasal function.
Objective assessment of nasal function included nasal peak inspiratory flow (NPIF), active anterior rhinomanometry, and acoustic rhinometry, all validated in patients undergoing rhinoplasty.34 Measurements were taken in a seated position after the patient had rested for at least 10 minutes in a climate-controlled room (22°C).
Nasal peak inspiratory flow assessed collapsibility using an In-Check Nasal inspiratory flow meter (Clement Clarke International). A tight seal was applied without compressing the external nares. The patient performed a maximal forced inspiratory effort through the nose with the mouth closed. The best of 3 attempts was recorded to accommodate patient learning.35,36 Values within 10% of each other were accepted as adequate patient technique.
Nasal airway resistance (NAR) was measured using 4-phase active anterior rhinomanometry (A6 Rhinomanometer; GM Instruments) with a fixed reference level of 150 Pa according to international standards.37 The patient held an anesthetic airtight mask around the nose, with the contralateral nostril sealed with a nasal plug. The patient was instructed to breathe normally through the nose with the mouth closed. The opposite side was then tested using the same method. The process was repeated until 2 consistent measurements were produced within 15% of each other. Representative values from each side were combined by use of NARIS software (GM Instruments) to obtain total NAR.
The minimum cross-sectional area (MCA) was measured with acoustic rhinometry (GM Instruments). The sound tube was applied to the nostril using an appropriately sized nosepiece to create an airtight seal. Patients were instructed to hold their breath during each measurement, which consisted of 4 readings.38 This process was repeated until 2 consistent MCA results were produced within 15% of each other. The MCA was measured similarly on the other side.
Statistical analysis was performed with SPSS, version 22 (SPSS Inc). Continuous data had normality assessed using frequency histograms. Independent 2-sided t tests compared patient age, body mass index, VAS score, NOSE score, SNOT22 score, physical component summary of the 36-item Short Form Health Survey version 2, NPIF, NAR, and MCA. The Kendall tau-b statistic was used to compare obstruction and nasal function scores, and the χ2 or Fisher exact test was used for sex and revision subgroup analysis. P < .05 was considered statistically significant.
A total of 495 patients (mean [SD] age, 36.5 [13.6] years; 302 women [61.0%] and 193 men [39.0%]; mean [SD] body mass index [calculated as weight in kilograms divided by height in meters squared], 23.6 [4.1]) were assessed, with 132 (26.7%) having had prior rhinoplasty. All patients completed the MCS, and 202 patients (mean [SD] age, 36.5 [14.3] years; 110 women [54.5%] and 92 men [45.5%]; mean [SD] body mass index, 23.9 [4.4]), with 28 (13.9%) having had prior rhinoplasty, completed all 3 questionnaires, which were added at a later date to explore other domains of mental health. Groups considered to be mentally healthy were determined according to the following 3 different categories: those with healthy mental well-being (MCS), recorded for 350 patients (70.7%); those with healthy self-esteem (RSES), recorded for 188 of 202 patients (93.1%); and those with healthy dysmorphic concerns (Dysmorphic Concerns Questionnaire), present in 156 of 202 patients (77.2%). The baseline characteristics for patients with poor mental health and for patients with good mental health were similar when defined by MCS and RSES scores (Table 1 and Table 2). Importantly, the proportions of patients undergoing a rhinoplasty revision were similar between the groups, making this variable less likely to be a confounder, and no differences were found on subgroup analysis of patients who underwent a revision. However, those with high levels of dysmorphic concerns had a higher revision rate (16 of 46 [35.0%] vs 28 of 156 [8.0%]; P = .04) and were 5 years younger (mean [SD] age, 32.3 [11.9] vs 37.8 [14.8] years; P = .03) (Table 3). Further analysis of the differences between patients in the group with dysmorphic concerns who underwent revision did not reach significance; however, this finding may be due to small population numbers. The outcomes between patients who underwent prior rhinoplasty and those who were surgery naive are defined in Table 4, with no clinically significant differences in patient-reported outcome measures or nasal airflow analysis outcomes.
Patients with poor mental well-being perceived their nasal function as significantly poorer in all patient-reported outcome measures (Table 1). Specifically, those with poor mental health had significantly poorer mean (SD) scores than patients with good mental health in the sensation of nasal obstruction using the VAS for the left side (51  vs 42 ; P = .001), VAS for the right side (54  vs 45 ; P < .001), NOSE (2.64 [0.95] vs 1.96 [1.04]; P < .001), and SNOT22 (2.14 [0.84] vs 1.33 [0.83]; P < .001). This difference continued with the nasal obstruction symptom more often rated as severe or worse in those with poor mental well-being (58 of 145 [40.0%] vs 83 of 350 [23.7%]; P < .001). A similar association existed for the nasal function anchor score more commonly being rated as poor or worse in patients with poor mental well-being vs patients with good mental health (72 of 145 [49.7%] vs 111 of 350 [31.7%]; P < .001). Objectively, nasal function was clinically similar between patients with poor mental well-being and those with good mental health (Table 1). While mean (SD) NPIF was significantly different between the groups (107  L/min in those with poor mental well-being vs 119  L/min in those with good mental health; P = .006), this difference was less than the minimum clinically important difference of 20 L/min.36 Further supporting the similarity in nasal function between the patients with poor mental well-being and those with good mental health were the other parameters of NAR (mean [SD], 0.519 [0.44] vs 0.532 [1.00] Pa/cm3/s; P = .89) and MCA (mean [SD], 0.80 [0.33] vs 0.85 [0.32] cm2; P = .20).
Overall, patients with low self-esteem perceived their nasal function as poorer compared with that of patients with healthy self-esteem (Table 2). Although there was no difference between patients with low self-esteem and those with healthy self-esteem in mean (SD) VAS score for the left side (42  vs 43 ; P = .85) or mean (SD) VAS score for the right side (57  vs 44 ; P = .09), those with low self-esteem had significantly poorer mean (SD) NOSE scores (3.03 [0.96] vs 2.04 [1.11]; P = .001) and mean (SD) SNOT22 scores (2.27 [0.85] vs 1.42 [0.91]; P = .001). The nasal obstruction symptom was rated severe or worse by more patients with low self-esteem (5 of 14 [35.7%] vs 50 of 188 [26.6%]; P = .009). The nasal function anchor score was also rated as poor or worse by significantly more patients with low self-esteem than those with good mental health (100% vs 112 of 188 [59.6%]; P = .004). Airflow analysis showed that the mean (SD) NPIF was significantly poorer in those with low self-esteem than those with healthy self-esteem (109  vs 135  L/min; P = .03). This clinically important difference of 26 L/min may be a product of the small sample sizes (n = 14), and a higher-powered study would be better able to confirm this finding. Given that other objective outcomes of nasal function were similar between those with poor self-esteem and those with healthy self-esteem (mean [SD] NAR, 0.404 [0.27] vs 0.464 [1.16] Pa/cm3/s; P = .85; mean [SD] MCA, 0.75 [0.31] vs 0.84 [0.33] cm2; P = .33), it is suggestive that the NPIF results were owing to chance rather than to a difference in nasal function between the groups.
Dysmorphic concerns did not have an association with any patient-reported outcome pertaining to nasal function (Table 3). There was no difference between patients with low levels of dysmorphic concerns and those with high levels of dysmorphic concerns in mean (SD) VAS score for the left side (43  vs 44 ; P = .80), mean (SD) VAS score for the right side (45  vs 46 ; P = .76), mean (SD) NOSE score (2.08 [1.13] vs 2.21 [1.12]; P = .48), or mean (SD) SNOT22 score (1.42 [0.91] vs 1.70 [0.96]; P = .07). In addition, there was no difference between patients with low levels of dysmorphic concerns and those with high levels of dysmorphic concerns in nasal obstruction symptoms (91 of 156 [58.3%] vs 21 of 46 [45.7%]; P = .56) or nasal function anchor scores (96 of 156 [61.5%] vs 30 of 46 [65.2%]; P = .49). Nasal function was objectively similar between patients with good mental health and those with high levels of dysmorphic concerns in mean (SD) NPIF (135  vs 126  L/min; P = .21), mean (SD) NAR (0.464 [1.27] vs 0.448 [0.27] Pa/cm3/s; P = .94), and mean (SD) MCA (0.83 [0.32] vs 0.82 [0.37] cm2; P = .91).
Despite mental health being a focus of many rhinoplasty studies, its association with patient perceptions of nasal function has not been well characterized. This study found that, preoperatively, patients undergoing rhinoplasty who had poor mental well-being as defined by the MCS and low self-esteem as defined by the RSES had significantly poorer perceptions of nasal function compared with those who were mentally healthy. Overall, the results of airflow analysis suggest that these groups had comparable objective nasal function at a clinically important level. Although mean (SD) NPIF was significantly different between those with poor mental well-being and those with good mental health (107  vs 119  L/min; P = .006), the absolute mean (SD) difference of 12 [4.5] L/min was less than the minimum clinically important difference of 20 L/min.36 The other nasal obstruction parameters of mean (SD) NAR (0.519 [0.44] vs 0.532 [1.00] Pa/cm3/s; P = .89) and mean (SD) MCA (0.80 [0.33] vs 0.85 [0.32] cm2; P = .20) were similar between those with poor mental well-being and those with good mental health. However, when patients were grouped according to their level of self-esteem, a difference emerged in NPIF, with patients with low self-esteem having poorer nasal function. Unfortunately, the small group size (n = 14) makes it challenging to better characterize this observation. It is possible that this result is a type 1 error, especially given that the other objective outcomes of nasal function (NAR and MCA) were similar between the groups. An alternative explanation may be that NPIF requires more patient effort compared with NAR and MCA, and patients with lower self-esteem may be less motivated to participate in nasal airflow testing. A subgroup analysis of those with low self-esteem comparing only NPIF with patient-reported NOSE scores found no correlation between the 2 measures (Spearman ρ = −0.067; P = .35), making it more likely that self-esteem confounds perception of nasal function. However, to be more confident of this phenomenon, again, a larger-scale study would need to be conducted. Thus, overall, the similarity observed in objective analysis of airflow lends confidence to the direction of the association that mental health influences perception of nasal function rather than poor nasal function adversely affecting mental health; however, a larger-scale study would lend more confidence to these findings.
Clinical wisdom supports the idea that patients with dysmorphic concerns focus primarily on appearance, but it was previously unknown as to whether this idea was associated with perceptions of function. We demonstrated that poor perceptions of nasal image are not transferred to beliefs of nasal function, which was demonstrated by the lack of significant difference between the patients with high levels of dysmorphic concerns and the patients with low levels of dysmorphic concerns. Hence, we have demonstrated that impairments in general mental well-being and self-esteem, but not dysmorphia, may translate to a negative self-evaluation of nasal function and influence the motivation for surgery.
Although society and the body of research often dichotomize rhinoplasty as functional or cosmetic, surgeons acknowledge that rhinoplasty is an operation requiring attention to both function and appearance.39-41 Furthermore, it has been acknowledged that there is often a fine distinction between cosmetic and medical indications for surgery,42 with some patients presenting with functional issues and not disclosing cosmetic motives, possibly owing to the stigma associated with cosmetic surgery.15 Conversely, patients seeking rhinoplasty primarily for cosmetic reasons may also have underlying functional problems. This scenario was illustrated in a previous study showing higher VAS and SNOT22 scores for 166 candidates for cosmetic rhinoplasty compared with population controls; 62% of patients had at least 1 structural sinonasal pathologic condition noted on clinical endoscopic examination.23 Patients’ motives were not assessed to allow separation between those presenting for only cosmetic changes and those with only functional concerns. However, in practice, patients uncommonly present with concerns for only cosmesis or function and usually have some combination of both. We acknowledge that patients’ motives will be practice dependent.
This study has some limitations. These findings relate to an Australian population of patients undergoing rhinoplasty, and generalizability to patients in other countries is limited owing to different cultural norms regarding mental health and perceptions of function.13,43 Considering the increasing popularity of rhinoplasty in Iranian, Asian, and Hispanic populations,7,43 it is worthwhile to explore whether these findings translate cross-culturally.
Observational studies are necessary in the context of rhinoplasty or mental health when the disease state or treatment options cannot be controlled by the researcher and when randomization and blinding are not feasible.44 The cross-sectional nature of this study minimized participant bias because nasal airflow analysis is a routine part of preoperative care for patients undergoing rhinoplasty. Therefore, all of the consecutive patients completed patient-reported outcomes and had their nasal function assessed. However, because this was a preoperative nasal airflow analysis, not all of the patients then underwent rhinoplasty at our centers. It is difficult for us to determine if the patients who did not undergo rhinoplasty at our center then decided not to have the operation or had surgery elsewhere. Bias was reduced by following standardized protocols of data collection and interpretation of nasal function outcomes and mental health assessments.
A strength of this study was the ability to compare patient-reported outcomes with different parameters of nasal function. These parameters included the NPIF, measuring the maximal nasal flow a patient can achieve during a maximal forced inspiration, which is better able to assess collapsibility with inspiration36; the NAR, assessing nasal airway resistance39; and the MCA, assessing nasal airway dimensions.38 The discrepancy between our study’s NPIF and NAR and MCA results may be due to differences in these parameters but is more likely to be due to the statistically significant difference in NPIF not translating to a clinically important level (as already described). At the same time, these assessments of nasal function have attracted criticism as being limited in their replication, being thought impractical in the clinical environment, requiring specialized equipment and a well-trained operator, and being time intensive.38,39,45 However, the NPIF is practical and reliable and, with careful interpretation for a clinically important difference,36 can effectively assess nasal function and may be a useful tool to integrate into a preoperative rhinoplasty evaluation.
For the surgeon, evaluating a patient’s mental health status can be challenging within the constraints of preoperative consultations. Because mental health is a complex and subjective concept, it is challenging to measure.46 It is common practice for surgeons to rely on “instinct” and “experience” when selecting patients,21 but this subcognitive judgment could be aided with patient self-reports. Although there is no perfect way to measure mental health, questionnaires such as the MCS of the 36-item Short Form Health Survey version 2 are useful in the surgical setting because they are validated, short, and able to be administered by a nonspecialist.42 This aspect is important because many other rhinoplasty studies have relied on assessment by a psychiatrist,4,10,11,47 but mental health professionals are rarely involved in the surgical environment. Although validated mental health questionnaires may be practical in a surgical context, they are limited by relying on the accuracy of a patient reporting his or her own attitudes, values, and feelings, which are often complicated by intended or unintended motives for self-presentation.27 Questionnaires, however, should be interpreted as screening tools that flag at-risk patients for a more thorough clinical assessment. Hence, such screening questionnaires are a practical aid but do not replace clinical judgment of mental health.
These results provide rhinoplasty surgeons and other medical professionals with the understanding that a patient’s mental health status may be associated with his or her perception of nasal function. As such, patients with poor mental health status presenting with functional concerns should be treated cautiously. For instance, additional consultations with the surgeon or physician, referral for objective nasal airflow assessment or assessment by NPIF, and possibly conjoint psychological therapy42 may clarify the clinical picture. Ethically, this additional treatment would facilitate better support of vulnerable patients with poor mental health and enhance their informed consent48 as to whether the operation is necessary from a functional perspective. Finally, having a more accurate understanding of the preoperative function of a patient considering rhinoplasty can enhance surgical planning and decision making.
Differences in mental health status, defined by practical and acceptable questionnaires, can be associated with patient perceptions of nasal function. Preoperatively, patients undergoing rhinoplasty who had poor mental well-being and low self-esteem reported poorer perceptions of nasal function at baseline compared with patients with good mental health despite having similar nasal airflow analysis outcomes. Thus, self-reports from a patient with a poor mental health status may not reflect true nasal function and, hence, should be viewed with caution by the surgeon.
Corresponding Author: Erika Strazdins BS Hons(Med), St Vincent’s Centre for Applied Medical Research, University of New South Wales, 405 Liverpool St, Sydney, New South Wales 2011, Australia (email@example.com).
Accepted for Publication: February 16, 2017.
Published Online: July 20, 2017. doi:10.1001/jamafacial.2017.0459
Author Contributions: Ms Strazdins 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.
Study concept and design: Strazdins, Christensen, Harvey.
Acquisition, analysis, or interpretation of data: Strazdins, Nie, Ramli, Palesy, Christensen, Marcells.
Drafting of the manuscript: Strazdins.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Strazdins, Nie, Christensen, Harvey.
Obtained funding: Harvey.
Administrative, technical, or material support: Strazdins, Palesy, Christensen, Marcells, Harvey.
Study supervision: Marcells, Harvey.
Conflict of Interest Disclosures: Dr Harvey reported serving as a consultant with Medtronic, Olympus, and NeilMed Pharmaceuticals; receiving research grant funding from Meda Pharmaceuticals and Stallergenes; and serving on the speakers bureau for GlaxoSmithKline and Arthrocare. No other conflicts were reported.
Create a personal account or sign in to: