Pain Catastrophizing and Quality of Life in Adults With Chronic Rhinosinusitis

Abstract

Objectives/Hypothesis:

Psychological comorbidity is common in patients with chronic rhinosinusitis (CRS) and is correlated with decreased overall and disease-specific quality of life (QoL). Prior research reported that anxiety and depression, as measured by the hospital anxiety and depression scale (HADS), are associated with worse CRS-specific QoL, as assessed via the Rhinosinusitis Disability Index (RSDI). Furthermore, patients prone to anxiety/depression may display an exaggerated response to real or anticipated discomfort; the pain catastrophizing scale (PCS) is a validated instrument designed to measure this phenomenon. This study is intended to explore the role of pain catastrophizing in relation to anxiety, depression, and disease-specific QoL in patients with facial pain attributed to CRS.

Study Design:

Prospective cohort study.

Methods:

Diagnosis of presumed CRS was based upon current American Academy of Otolaryngology - Head & Neck Surgery (AAO-HNS) guidelines; all participants reported facial pain as a component of their CRS symptomatology. RSDI, HADS, and PCS questionnaires were administered upon presentation prior to intervention, and objective measurements of sinonasal inflammation were obtained via nasal endoscopy and computed tomography (CT).

Results:

Seventy-five patients were enrolled in the study. Significant positive correlations were found between PCS and HADS, total RSDI, and RSDI emotional sub-scores (P < .05). The incidence of objective evidence of disease, as measured via nasal endoscopy and CT, was not significantly different in catastrophizing patients.

Conclusions:

Pain catastrophizing correlates with anxiety/depression and worse disease-specific QoL in patients meeting symptomatic criteria for CRS. Otolaryngologists should be aware that catastrophic thinking can intensify a patient’s perception of sinonasal symptoms, and clinicians may consider management of psychological comorbidity to optimize rhinologic outcomes.

INTRODUCTION

Chronic rhinosinusitis (CRS) is a prevalent condition that can lead to a range of clinical manifestations, with considerable impact on quality of life (QoL). As reflected in current diagnostic criteria, cardinal symptoms of CRS include nasal congestion, anterior or posterior rhinorrhea, hyposmia, and facial pain.¹ Pain may be a prominent component of the disease for many patients, and prior diagnostic approaches have identified headache, dental pain, and ear pain/pressure/fullness as additional symptoms associated with CRS. The variety of pain included in older diagnostic criteria highlights the considerable range of head and neck discomfort possible in this disorder.² In patients with objective evidence of CRS in the form of abnormal nasal endoscopy or computed tomography (CT) scan, 16% to 20% complained of pain as a component of their disease.³

Despite the high prevalence and significant public health burden associated with CRS, symptomatology is surprisingly imprecise in predicting the presence and magnitude of objective evidence of sinonasal inflammation, with multiple studies demonstrating a relatively poor correlation between symptom-based diagnostic criteria and abnormalities on CT or nasal endoscopy.^4–10 Prior research has demonstrated pain to have a poor correlation with objective sinonasal disease, with a variety of other medical conditions including temporomandibular joint disorder (TMJD), migraine, myofascial syndromes, and atypical facial pain capable of producing similar nociceptive symptoms.³ Despite limitations regarding the clinical utility of pain in the diagnosis of CRS, recent data suggest many patients consider this symptom to be important.¹¹

An abundance of research has explored the relationship between psychiatric comorbidity and CRS, with an increased prevalence of depression and anxiety noted in patients with symptoms attributed to rhinosinusitis.¹² Furthermore, individuals with CRS and comorbid psychological disorders have been noted to have worse disease-specific and overall QoL compared to patients with CRS but no psychological comorbidity.^9,12 In a study by Tomoum et al, increased depression and anxiety, as measured by the Hospital Anxiety and Depression Scale (HADS), were found to correlate with worsened disease-specific QoL, indicated by higher values on the Rhinosinusitis Disability Index (RSDI).^9,13

Medical therapy is the mainstay of initial treatment of CRS, followed by surgical intervention for refractory disease. Despite aggressive management, many patients with CRS continue to report decreased QoL that imperfectly correlates with objective evidence of disease upon CT or nasal endoscopy.⁴ Recent studies have related the discrepancy between subjective impairment and objective disease burden to external factors, including age, cultural expectations, and comorbid conditions, notably psychiatric illnesses such as depression and anxiety.^7,9,10,14 Studies exploring the effects of surgical treatment on QoL in patients with comorbid depression and anxiety have produced variable results, with some reporting a similar magnitude of QoL improvement following surgery when compared to the overall CRS population, but others showing a reduced level of benefit.¹²

The accentuated impact of chronic illness on QoL observed in patients with depression and anxiety may be related to increased awareness of and perseveration regarding disease symptoms. Multiple studies have identified a heightened perception and exaggerated response to real or anticipated stimuli known as pain catastrophizing (PC). Pain is a complex phenomenon, with a patient’s experience of discomfort shaped by an interaction between physical stimuli and psychological factors. Highly correlated with depression and anxiety, PC has been demonstrated to negatively impact QoL in a number of chronic diseases, including endometriosis, TMJD, chronic pelvic pain, inflammatory bowel disease, fibromyalgia, and sickle cell disease.^15–20 Despite the frequency with which CRS patients report pain as a primary symptom of their disease, to our knowledge the role of PC has not been studied in relation to CRS.

The goal of the current study was to determine the relationship among PC, psychological comorbidity, and CRS-specific QoL. We hypothesized that PC would positively correlate with the presence of anxiety and depression, as well as worse CRS-specific QoL, but would not predict objective evidence of sinonasal disease. Furthermore, PC was assessed in relation to certain comorbidities characterized by an accentuated presence of facial pain, including migraine, TMJD, and fibromyalgia.

MATERIALS AND METHODS

This cross-sectional study was approved by the Institutional Review Board at Virginia Commonwealth University. Adults aged 18 years and older with symptoms consistent with CRS according to current American Academy of Otolaryngology guidelines were included. All patients reported >3 months of facial pain plus at least one other cardinal symptom of CRS (nasal obstruction, anterior or posterior rhinorrhea, or anosmia). Exclusion criteria included pregnancy, mental disability, current illicit drug use, sinonasal neoplasm, and incarceration. Enrollment was capped at 75 patients. Demographic data were noted in chart review, as well as relevant medical history, specifically the presence of other pain related comorbidities including migraine, anxiety, depression, fibromyalgia, and TMJD.

Patients completed a variety of questionnaires including RSDI, HADS, and a pain catastrophizing scale (PCS) at the initial visit. RSDI is a survey divided into three components (functional, emotional, and physical) used to assess QoL in patients with CRS.¹³ The HADS is a validated, self-reported survey used to determine clinically significant states of anxiety and depression in a non-psychiatric hospital setting.^9,21 The PCS is a validated 13-item Likert scale questionnaire initially developed in 1995 by Michael Sullivan that is able to assess the impact of catastrophizing on pain experiences. Participants are asked to rate pain experiences on a scale from 0 to 4 based on their feelings when experiencing the pain; with a maximum score of 52, higher scores indicate a higher propensity for catastrophic thinking in relation to pain.²² Nasal endoscopy was quantified using the Lund-Kennedy (LK) endoscopic scoring system which evaluates each nasal cavity for the severity of polyps, edema, discharge, scarring, and crusting.²³ For the purposes of our study, a score greater than three was considered clinically significant. Patients were treated with medical management for CRS according to current International Consensus Statement on Rhinosinusitis (ICAR-CRS) guidelines, including (unless contraindicated) oral antibiotics, high volume nasal saline lavages, and topical nasal corticosteroids.²⁴ Following maximal medical treatment of at least 4 weeks duration, patients underwent CT scanning, with results scored according to the Lund-Mackay (LM) system, a method of standardizing the assessment of CRS in patients undergoing endoscopic sinus surgery by grading the opacification for each of the sinuses as well as the osteomeatal complex.²⁵ A score greater than three was considered clinically significant for the purpose of this study.²⁶

Data were summarized using means and standard deviations or frequencies and percentages. Pearson’s correlations were calculated to assess the relationship between PC and dysfunction and disability. Fisher’s transformation was used to estimate confidence intervals.

RESULTS

A total of 75 patients were prospectively enrolled in the study, with an average age of 47.6 ± 14.3 years (range 19–80 years). Sixty-nine percent of patients were female, and 24% had a history of functional endoscopic sinus surgery (FESS). A clinical history of anxiety or depression was reported in 12% of patients, whereas 53% had a history of a pain-related comorbidity. Objective evidence of sinonasal disease was seen in 71% (n = 53) of patients: 37% (n = 28) had abnormal nasal endoscopy (LK > 3) and 41% (n = 31) an abnormal CT (LM > 3). Twenty-one percent of patients (n = 16) had both abnormal endoscopy and CT. Complete patient characteristics are listed in Table I.

TABLE I.

Demographics of the Study Cohort.

Age (years), mean (SD)	47.6 (14.3)
Sex, n (%)
Male	23 (31%)
Female	52 (69%)
Previous FESS, n (%)	18 (24%)
Comorbidities, n (%)
Asthma	19 (25.3%)
Migraine	28 (37.3%)
Anxiety/Depression	9 (12%)
Fibromyalgia	3 (4%)
Chronic pain	3 (4%)
Objective evidence of sinonasal disease, n (%)
Abnormal nasal endoscopy (LK > 3)	28 (37%)
Abnormal CT scan (LM > 3)	31 (41%)
Anxiolytic/Antidepressant use, n (%)	27 (36%)

CT = computed tomography; FESS = functional endoscopic sinus sur-gery; LM = Lund McKay; LK = Lund Kennedy.

Table II displays mean and standard deviation scores for the total RSDI, HADS, and PCS, as well as for their individual sub-scores, for patients with and without objective evidence of sinonasal disease. There were no differences in mean scores for any of these instruments when patients with and without objective evidence of disease were compared (P > .05). Furthermore, no differences existed in RSDI, HADS, and PCS between patients with and without comorbid conditions.

TABLE II.

Mean Scores of Questionnaires Stratified by Presence of Objective Evidence of Disease.

	No Disease*		Disease*		Correlation With Total PCS
	Mean	SD	Mean	SD	r	P Value
RSDI
Total	36.3	27	43.5	25.7	0.638	<.001
Emotional	10.7	10.8	12.1	10	0.605	<.001
Functional	9.1	8.3	12.8	9.1	0.626	<.001
Physical	16.6	11.3	18.2	9.8	0.506	<.001
HADS
Total	13.6	7.7	14.8	8.1	0.705	<.001
Anxiety	7.6	4.6	8.3	4.2	0.647	<.001
Depression	6	3.9	6.5	4.5	0.644	<.001
PCS
Total	17.5	13	20.2	14.8	–	–
Rumination	6.1	4.7	6.5	4.9	–	–
Magnification	3.8	3.3	4.3	3.3	–	–
Helplessness	7.4	6	9.4	7.1	–	–

^*Rhinosinusitis Disability Index (RSDI), hospital anxiety and depression score (HADS), and pain catastrophizing scale (PCS) scores did not significantly differ between the disease and no disease groups (P > .05). Disease defined as either Lund McKay (LM) or Lund Kennedy (LK) score greater than or equal to 3.

PCS scores were positively associated with total RSDI (r = 0.638, P < .001) as well as the functional, emotional, and physical sub-scores of the RSDI instrument (see Table II and Fig. 1). Correlations between PCS and RSDI scores ranged from 0.51 to 0.63 without adjusting for sex, age, and emotional dysfunction, and 0.36 to 0.43 when adjusting for age, sex, and disability. Furthermore, there were strong correlations between the total HADS scores, as well as the anxiety and depression sub-scores of the HADS instrument, and PCS (R = 0.705, P < 0.001) (see Table II and Fig. 2). Unadjusted correlations of the PCS with the HADS were around 0.65 for each of the anxiety and depression subscales and 0.71 for the total score. These values ranged from 0.47 to 0.56 after adjusting for the aforementioned variables. Finally, a significant positive correlation was noted between total HADS and total RSDI (see Fig. 3).

Fig. 1.

Correlation of Pain Catastrophizing Scale (PCS) with total Rhinosinusitis Disability Index (RSDI) and RSDI sub-scores.

Fig. 2.

Correlation of Pain Catastrophizing Scale (PCS) with total Hospital Anxiety and Depression Score (HADS) as well as HADS anxiety and depression sub-scores.

Fig. 3.

Correlation of total Rhinosinusitis Disability Index (RSDI) with total Hospital Anxiety and Depression Score (HADS).

DISCUSSION

CRS is a prevalent condition that, despite aggressive medical and surgical therapy, is commonly associated with poor overall and disease-specific QoL.⁴ In addition to significant sinonasal symptomatology, numerous studies have demonstrated diverse systemic effects of CRS, including decreased sexual function, fatigue, and other psychosocial symptoms.^27,28 The impact on QoL due to CRS has been reported to be greater than that associated with other chronic diseases such as congestive heart failure and chronic obstructive pulmonary disease, demonstrating the significant nature of this problem.²⁹

Current research has explored the relationship between psychiatric comorbidities, specifically anxiety and depression, and patient-reported QoL in multiple chronic diseases.¹⁰ With regard to CRS, an elevated prevalence of anxiety and depression have been reported, and disease-specific QoL is decreased in patients with these psychological comorbidities.⁹ This association between psychiatric disease and impaired CRS-specific QoL was replicated in the current study, with a significant and positive correlation noted between total HADS and RSDI (see Fig. 1). In addition to correlating with decreased overall and disease-specific QoL at baseline, anxiety and depression have been noted in multiple studies to portend varying degrees of impaired QoL following surgical treatment.^4,12,30

Patients with depression have a higher prevalence of pain than the general population.³¹ As pain is a cardinal symptom of CRS, it follows that depressed patients with CRS would be more likely to have a significant burden of pain-related QoL impairment.^3,11,31–33 Although the mechanism of this interaction is complex and has not been fully elucidated, some studies have suggested a positive feedback loop between pain and emotional angst that leads to a cycle of worsening disability. To this end, Sullivan et al reported that accentuated pain perception in the form of catastrophic thinking resulted in higher levels of emotional distress, as well as an increased likelihood that the pain would persist over time.²²

Our study demonstrated that PC behavior correlated with objective measures of anxiety and depression (see Table II). The relationship between catastrophic thinking and psychological comorbidity was noted by elevated overall HADS score as well as for its individual measures of anxiety and depression (see Fig. 2), suggesting that this effect is not unique to either of these psychiatric diagnoses. Furthermore, the data demonstrated that in patients with CRS who reported facial pain as a component of their symptomatology, PC was associated with worse disease-specific QoL (see Table II). As illustrated in Fig. 3, this relationship remained significant when total PCS scores were correlated to the individual functional, physical, and emotional RSDI sub-scores. These findings suggest that the effects of catastrophic thinking are broad with regard to CRS-specific QoL, not limited to the patient’s perception of the emotional aspects of their disease.

The burden of PC in chronic disease is significant, raising the question of whether treatment aimed at these behaviors can positively impact QoL. Several studies have demonstrated that early recognition of catastrophic thinking, with management through appropriate intervention, can result in reduction of subjective levels of pain and disability.^22,34 Research has suggested that not only is PC modifiable, but that treatment to decrease catastrophic thinking can result in overall improved clinical outcomes.²² Jenson et al reported that for patients with chronic pain, an outpatient therapeutic program that provided pain-management skills in addition to physical and psychological therapy resulted in improved pain-related thinking and coping mechanisms. Patients reported a perception of increased control over chronic discomfort, thereby decreasing the disability associated with their symptomatology.³⁵

In addition to psychological intervention, additional treatment modalities have shown benefit in the management of patients with various diseases exhibiting PC behavior. Smeets et al employed physical therapy, cognitive behavioral therapy, or a combination of the two in patients with chronic low back pain, targeting the characteristic fear-avoidance behavior through positive experiences. These interventions improved both PC and QoL scores.³⁶ Further research would be useful to elucidate whether analogous psychological and physiotherapeutic approaches could be useful in improving QoL in CRS patients with chronic facial pain who exhibit PC behavior.

A recent study by Adams et al reported that anxiety and depression remain unchanged following FESS.⁴ This observation suggests the perseverance of psychological comorbidity despite aggressive management of sinonasal disease, and the need for ongoing clinician awareness of the effects of anxiety and depression throughout the perioperative period. A team-based approach, including a mental health provider, may offer advantages in the management of CRS patients with comorbid anxiety and depression, particularly when QoL impairment is severe and PC behavior is present.

This study has several potential limitations. Gender has been shown to have an impact on the perception of disease burden, as women are not only more likely to report facial pain but also to rate increased severity.³⁷ Given that the current study sample was composed of 69% women, the results may have skewed the cohort towards an overall worse QoL due to higher levels of pain symptomatology. Arguing against a substantial influence of gender in the results, however, was the fact that subgroup analysis revealed no differences between men and women in any of the endpoints measured. Nevertheless, while the presence of facial pain was evaluated in this study, the intensity and character of the pain were not explored, and may benefit from future research.

A second potential limitation of the current study was the reliance on self-reported anxiety and depression rather than formal evaluation by a clinician with experience in psychiatric diagnosis. While symptom scores on the HADS have been demonstrated to strongly correlate with a Diagnostic and Statistical Manual of Mental Disorders 5th edition (DSM-V) diagnosis of anxiety or depression, there remains the possibility of under-reporting of clinically significant psychological disease.²¹ A disconnect between self-reported behavioral symptoms and a formal psychiatric diagnosis was evident in the current study sample, with the total HADS scores of 50 patients indicating the presence of anxiety or depression, despite the fact that only nine reported a medical history of either of these diagnoses. This discrepancy may have been due to a tendency of patients to under-report mental health history, or of under-identification of clinically relevant psychiatric disease within this tertiary rhinology population. These results suggest that the otolaryngologist should remain vigilant for evidence of anxiety or depression when treating the patient with symptomatic CRS, even if the individual denies a history of a diagnosed psychological disorder.

A third limitation of this study was its cross-sectional nature, which limited the ability to assess which measures were impacting others. A longitudinal study design would allow characterization of the temporal relationship between these variables, and should be considered in future research on the topic of PC and CRS-specific QoL.

Radiographic evidence was not obtained at the same time as the HADS, RSDI, and PCS surveys, which were completed at the initial office visit. Nasal endoscopy was performed at initial evaluation, providing an objective evaluation of sinonasal disease coincident to psychological assessment. Radiography, however, was not performed until after completion of additional medical therapy, in order to characterize persistent sinonasal inflammation. The temporal gap between the questionnaires and CT imaging could be argued to have reduced the reliability of correlations among these results. Conversely, persistence of objective disease on CT despite maximal medical therapy might argue that this was a particularly robust measure of clinically relevant sinonasal inflammation in this population. It is important to emphasize that PCS, HADS, and RSDI did not differ between groups with or without evidence of disease on either initial nasal endoscopy or subsequent CT scanning, suggesting the effects of timing were minimal.

The current study sample was composed of patients who pursued treatment by a fellowship-trained rhinologist at a tertiary, academic center, and thus the results may not be representative of a general population of patients with CRS. A significant proportion of patients in this cohort were likely to have already received treatment by an otolaryngologist, as indicated by the fact that nearly one quarter of patients (24%) had a history of prior sinus surgery. As such, the sample may have been biased towards individuals with recalcitrant sinonasal disease, or perhaps those with more significantly impaired QoL. A larger study with a more diverse patient population could be helpful to confirm whether the relationship between anxiety, depression, PC, and CRS-specific QoL can be extended outside of the tertiary care setting.

Of patients in the current study who met symptom-based diagnostic criteria for CRS, 29% did not have evidence of objective disease on subsequent CT or nasal endoscopy. This finding supports the body of literature demonstrating a relative disconnect between CRS symptomatology and objective testing for sinonasal disease. Nevertheless, it is important to understand the relationship between psychological comorbidity, symptoms, and QoL in this subset of patients, all of whom attributed their symptoms to sinusitis and in most cases had been diagnosed and treated for CRS by other clinicians. One possible explanation for this sizeable minority of symptomatic patients lacking objective evidence of sinonasal disease is that mucosal inflammation resolved in the interval between initial evaluation and CT, the latter of which was not performed on the same day as first clinic visit. Alternatively, these patients may have had low-level objective sinus disease not classified as abnormal by the criteria employed by this study, that is, LK > 3 or LM > 3. A third possibility is that patients had sinonasal symptoms, including facial pain, that were attributed to sinusitis but actually caused by a separate diagnosis, such as migraine, atypical facial pain, or TMJD. It is interesting to note, however, that the HADS, RSDI, and PCS scores, as well as objective measures of disease, did not differ between subsets of patients with and without a diagnosis of these pain-related comorbidities.

CONCLUSION

Anxiety and depression are common in patients with CRS and can impact the perception of pain, potentially manifesting as PC behavior, which is associated with a heightened perception and exaggerated response to painful stimuli and worse disease-specific QoL. Recognizing PC is critical to identification of those patients who may benefit from adjunctive multi-disciplinary management of anxiety and depression in order to optimize clinical outcomes.