Associations between fatigue and medication use in chronic rhinosinusitis.

Abstract

We conducted a prospective study of 586 adults to determine if associations exist between fatigue symptom scores and three classes of medications prescribed for the treatment of chronic rhinosinusitis (CRS): prescription nonsedating antihistamines, topical nasal steroids, and antibiotics. Patients were assessed with the assistance of the Rhinosinusitis Symptom Inventory and Likert-scale fatigue symptom scores. On multivariate analysis and correcting for disease severity, we found that significantly higher fatigue symptom scores were associated with the use of nonsedating antihistamines (mean Likert score: 2.75 vs. 2.27 for patients not taking a nonsedating antihistamine; p = 0.029). Higher fatigue scores were also associated with a greater number of antibiotic courses and more total weeks of antibiotic use (p < 0.001 in both cases). No association was seen between fatigue scores and the use of topical nasal steroids (mean Likert score: 2.65 vs. 2.24; p = 0.658). We recommend that long-term use of a nonsedating antihistamine be scrutinized in CRS patients who report symptoms of fatigue.

Introduction

Fatigue is a common symptom associated with the diagnosis of chronic rhinosinusitis (CRS). In fact, it is one of the criteria used to establish a diagnosis of CRS. Chronic fatigue places a significant quality-of-life burden on patients, (1,2) and it has attracted a significant amount of research effort because it is associated with many other chronic disease processes, such as infections, obstructive sleep apnea, and chronic fatigue syndrome itself. (3) We have often been struck by the frequency and degree of fatigue symptoms reported by our own patients who undergo evaluation for CRS. (4) Not infrequently, chronic fatigue is a significant factor that motivates patients to seek diagnosis and therapy for CRS.

The cause of fatigue in the CRS syndrome complex remains unknown. Symptoms of fatigue may be associated with the pathogenesis of CRS itself, with the psychological nature of the disease burden in CRS, and/or with the medication regimen often used to treat CRS. Many patients correlate their fatigue with the number of antibiotic courses they have taken. Other patients correlate fatigue with the use of nonsedating antihistamines, and others still with the use of topical nasal steroid sprays. In this article, we describe our study of the relationships between fatigue and medication use in CRS.

Patients and methods

We obtained data at the initial assessment of all adults who presented to our institution for evaluation of CRS. Patients were asked to complete the Rhinosinusitis Symptom Inventory to arrive at scores for major and minor CRS symptoms. (4,5) Patients were also asked to rate their fatigue symptomatology averaged over the previous 12 weeks on a 6-point Likert scale, ranging from 0 (no symptoms of chronic fatigue) to 5 (severe fatigue) in the form of an embedded question. For control purposes, patients were also asked to rate the severity of fever symptoms on the same 6-point scale. Fever is not a major symptom associated with chronic rhinosinusitis, and therefore it served as a good control comparator.

Embedded questions were used to ascertain past and current use of prescription nonsedating antihistamines, topical nasal steroids, and antibiotics. Antibiotic use was quantified as the number of courses and the number of weeks of oral antibiotic therapy during the preceding calendar year. Because over-the-counter antihistamines are well known to cause fatigue, these medications were excluded from this analysis (this study was conducted before loratadine became available as an over-the-counter preparation, and therefore use of this drug was included in the analysis). Patients were blinded as to the specific purpose of this study.

Data on symptoms and medication use were entered into a Microsoft Excel spreadsheet and analyzed for accuracy on a case-by-case basis. Data were then imported into SPSS version 10.0 software for statistical analysis. Standard demographic information with respect to gender, age, symptom scores, and medication use was computed. Next, one-way analysis of variance (ANOVA) was conducted with fatigue as the dependent variable and medication use as the factor variable influencing fatigue. A multivariate ANOVA model was then constructed with fatigue score as the dependent variable; current use of a nonsedating antihistamine and current use of a topical steroid were independent variables; the number of antibiotic courses was entered as a covariate. To control for the probability that increasing disease severity would make multiple medication use more likely (i.e., patients with more severe CRS are more likely to take both a nonsedating antihistamine and a topical nasal steroid), the combined nasal symptom score domain was also entered as a covariate. The nasal symptom domain score is an unweighted average of Likert scores for severity of symptoms for nasal obstruction, rhinorrhea, and dysosmia. A second statistical model was also run with the number of weeks of antibiotics (rather than the simple number of courses of antibiotics) as a covariate. Statistical significance was set at p < 0.05. A similar statistical analysis was conducted for fever symptom severity as the dependent variable in the multivariate model for purposes of control comparison.

Our protocol was approved by our hospital's committee on clinical investigations.

Results

A total of 586 patients provided complete data on symptom scores and medication use. The mean age was 40.3 years.

The mean fatigue score on the scale of 0 to 5 was 2.5 ([+ or -] 1.5) (figure). Only 16.2% of patients reported no element of chronic fatigue associated with their CRS.

[FIGURE OMITTED]

A total of 258 patients (44.0%) reported that they were taking a nonsedating antihistamine at the time of their evaluation; on average, antihistamines had been in use for 14.5 weeks during the preceding 12 months. Current use of a topical nasal steroid was reported by 346 patients (59.0%); the average duration of use (off and on) during the preceding 12 months was 16.6 weeks. The mean number of antibiotic courses prior to evaluation was 2.5 ([+ or -]2.2), and the mean number of weeks on antibiotic therapy was 5.3 ([+ or -]6.5).

On univariate analysis, higher fatigue scores were significantly associated with nonsedating antihistamine use (p < 0.001), topical nasal steroid use (p = 0.002), a greater number of antibiotic courses (p < 0.001), and more total weeks of antibiotic use (p = 0.004) (table). However, after correcting for disease severity, multivariate analysis revealed that fatigue was not significantly associated with topical nasal steroid use (mean Likert score: 2.65 vs. 2.24 for patients not taking a topical steroid; p = 0.658). Fatigue did remain significantly associated with antihistamine use (mean Likert score: 2.75 vs. 2.27 for patients not taking a nonsedating antihistamine; p = 0.029), a greater number of antibiotic courses (p < 0.001), and a greater number of weeks on antibiotics (p < 0.001). As expected, the nasal symptom domain score was positively correlated with increasing fatigue symptoms as a marker for increasingly severe symptoms of CRS and appropriately included in the model as a covariate (p < 0.001). An interaction term between antihistamine use and steroid use was not statistically significant (p = 0.807, ANOVA).

For the multivariate control comparison with fever symptoms, neither nonsedating antihistamines nor topical nasal steroids were associated with fever symptom severity (p = 0.309 and p = 0.932, respectively). The number of antibiotic courses did significantly correlate with fever scores (p = 0.001).

Discussion

The reason for the presence of fatigue symptoms in the setting of CRS remains an enigma. (3) The etiology is probably multifactorial, and it may be linked to the disease process of CRS itself, to the side effects of medication use, and/or to the psychosocial factors associated with the diagnosis of CRS. Specific associations between chronic fatigue, chronic rhinitis, and CRS have appeared on several occasions in the literature. (6) We previously identified a significant fraction of patients who reported moderate to severe symptoms of fatigue at the time of evaluation of CRS. (4) Conversely, Chester reported that CRS was more common in patients with chronic fatigue than in patients without symptoms of chronic fatigue. (7)

As further evidence of the link between fatigue and CRS, a quantitative evaluation of fatigue symptoms is often contained within disease-specific quality-of-life outcome measures for CRS, including the Rhinosinusitis Disability Index, the Sinonasal Outcome Tests, and the Rhinosinusitis Outcome Measure. (8) Several studies that incorporated more global health-related quality-of-life measures, including the Short Form 36, have also identified significant elements of fatigue in patients with CRS. (9,10)

Nonsedating antihistamines have long been a mainstay of therapy for the treatment of perennial rhinitis, allergic rhinitis, and CRS. Although none of the nonsedating antihistamines carries a specific Food and Drug Administration--approved indication for the treatment of CRS, they are widely used to control many of the symptoms potentially associated with CRS. Their prevalence in the treatment of CRS is evidenced by the substantial fraction of patients who have already used a nonsedating antihistamine prior to an otolaryngologic evaluation of CRS--44.0% in our study.

The introduction of nonsedating antihistamines represented a significant advance in the treatment of many allergic disorders, including allergic rhinitis and chronic urticaria. These antihistamines acquire their nonsedating pharmacologic profile through molecular modifications that limit diffusion across the blood-brain barrier. Their lack of central nervous system effects has been the subject of numerous investigations, most of which found that cognitive and motor skills are not significantly impaired by these medications, especially when compared with first-generation antihistamines. (11) However, chronic fatigue is a much more widespread clinical syndrome, and cognitive and motor skill disturbances represent only two components of it.

The clinical safety and efficacy of nonsedating antihistamines have been well established for many sinonasal conditions. Recently, in fact, loratadine was deemed safe enough to be accessed by patients as an over-the-counter medication. In our catchment area, this prompted many third-party payers to adopt over-the-counter nonsedating antihistamines as a first-line treatment for many nasal diagnoses, including rhinitis, allergic rhinitis, and even CRS. This new policy shifts the cost burden from the third-party payer to the patient as an out-of-pocket expense. In contrast, nasal steroids remain prescription medications, and they are often preferred as first-line agents for the management of CRS.

We have been somewhat concerned about the shift toward nonsedating antihistamines because of their potential to create or exacerbate fatigue symptoms despite being classified as "nonsedating." This shift affects patients with CRS because before a definitive diagnosis of CRS is made, many such patients are treated according to guidelines for allergic and perennial rhinitis. Furthermore, the impact of such a change in medication recommendations may be more significant for patients with CRS in whom fatigue already exerts a significantly negative influence on overall quality of life. These potential dangers may be additionally compounded in CRS by the fact that patients with CRS often have symptoms year-round and may therefore be subject to these fatigue-related side effects almost constantly, unlike seasonal users of nonsedating antihistamines.

Our study establishes an association between nonsedating antihistamines and fatigue symptoms associated with CRS, but not a causal association. A causal relationship can be established only by a prospective, controlled, randomized trial of nonsedating antihistamines in patients with CRS. However, we believe that our data do provide evidence that it is likely, at least to some degree, that some of the fatigue symptoms reported by patients with CRS are attributable to the use of nonsedating antihistamines. It is also possible that a synergistic effect exists between nonsedating antihistamines and other medications used to treat CRS, and that the latter may exacerbate the fatigue-related side effects of the former. Alternatively, it is also possible that a patient's fatigue is linked to an increase in the severity of CRS itself. However, if this were the case, one would also expect that topical nasal steroids would be associated with increasing fatigue scores on multivariate analysis. In fact, we attempted to control for increasing disease severity by including the nasal symptom domain score in the multivariate analysis, and therefore we believe the association between fatigue and the use of nonsedating antihistamines does exist. Finally, the magnitude of the difference in reported fatigue scores might be seen as small numerically. However, previous studies have determined that on 7-point rating scales, patients perceive a difference of approximately 0.5 units as clinically significant. (12)

Because chronic fatigue has a significant negative impact on quality of life for patients with CRS, we recommend that they be informed of the potential for additional fatigue symptoms before they take a nonsedating antihistamine. Patients should also be monitored for increasing fatigue throughout the course of therapy.

Because we did not find any association between fatigue symptoms and topical nasal steroids, these agents may be preferable as a first-line therapy for CRS. An oral nonsedating antihistamine could be added or substituted if symptoms persist or if side effects become unacceptable. But for now, the optimal medical treatment for patients with newly diagnosed CRS remains unknown and warrants continuing investigation.

References

(1.) Yeomans JD, Conway SP. Biopsychosocial aspects of chronic fatigue syndrome (myalgic encephalomyelitis). J Infect 1991;23:263-9.

(2.) Lanza DC, Kennedy DW. Adult rhinosinusitis defined. Otolaryngol Head Neck Surg 1997;117(3 pt 2):S1-7.

(3.) Shephard RJ. Chronic fatigue syndrome: An update. Sports Med 2001;31:167-94.

(4.) Bhattacharyya N. The economic burden and symptom manifestations of chronic rhinosinusitis. Am J Rhinol 2003;17:27-32.

(5.) Bhattacharyya N. Symptom outcomes after endoscopic sinus surgery for chronic rhinosinusitis. Arch Otolaryngol Head Neck Surg 2004;130:329-33.

(6.) Baraniuk JN, Clauw DJ, Gaumond E. Rhinitis symptoms in chronic fatigue syndrome. Ann Allergy Asthma Immunol 1998;81:359-65.

(7.) Chester AC. Symptoms of rhinosinusitis in patients with unexplained chronic fatigue or bodily pain: A pilot study. Arch Intern Med 2003;163:1832-6.

(8.) Leopold D, Ferguson BF, Piccirillo JF. Outcomes assessment. Otolaryngol Head Neck Surg 1997;117(3 pt 2):S58-68.

(9.) Gliklich RE, Metson R. The health impact of chronic sinusitis in patients seeking otolaryngologic care. Otolaryngol Head Neck Surg 1995;113:104-9.

(10.) Winstead W, Barnett SN. Impact of endoscopic sinus surgery on global health perception: An outcomes study. Otolaryngol Head Neck Surg 1998;119:486-91.

(11.) Simons FE. Antihistamines. In: Adkinson NF Jr., Yuninger JW, Busse WW, et al, eds. Middleton's Allergy: Principles and Practice. 6th ed. St. Louis: Mosby; 2003.

(12.) Guyatt GH, Juniper EF, Walter SD, et al. Interpreting treatment effects in randomised trials. BMJ 1998;316:690-3.

Neil Bhattacharyya, MD, FACS; Lynn J. Kepnes, RNP

From the Division of Otolaryngology, Brigham and Women's Hospital (Dr. Bhattacharyya and Ms. Kepnes), and the Department of Otology and Laryngology, Harvard Medical School (Dr. Bhattacharyya), Boston.

Reprint requests: Neil Bhattacharyya, MD, Division of Otolaryngology, 45 Francis St., Boston, MA 02115. Phone: [phone]; fax: [phone]; e-mail: [e-mail]

Table. Results of the univariate and
multivariate statistical analyses

                        Mean fatigue  Univariate   Multivariate
                          score *

                     not
Medication          using     Using    p value       p value

Nonsedating         2.27      2.75     < 0.001        0.029
  antihistamine
Topical nasal       2.24      2.65      0.002         0.658
  steroid
                  Correlation
                  coefficient

No. antibiotic      0.239              < 0.001       < 0.001
  courses

No. wk on           0.132               0.004        < 0.001
  antibiotics

Nasal symptom       0.319              < 0.001       < 0.001
  domain score

* Fatigue was rated on the Likert scale of O (none) to 5 (.severe).