Neuropathology in Rhinosinusitis
http://www.100md.com
美国呼吸和危急护理医学 2005年第1期
Division of Rheumatology, Immunology and Allergy, Georgetown University, Washington, DC
Department of Otorhinolaryngology, Kaohsiung Medical University, Kaohsiung City, Taiwan
Department of Otorhinolaryngology, Catholic University, Seoul, Korea
Department of Otorhinolaryngology, Mie University, Tsu, Japan
ABSTRACT
Pathophysiologic differences in neural responses to hypertonic saline (HTS) were investigated in subjects with acute sinusitis (n = 25), subjects with chronic fatigue syndrome (CFS) with nonallergic rhinitis (n = 14), subjects with active allergic rhinitis (AR; n = 17), and normal (n = 20) subjects. Increasing strengths of HTS were sprayed into their nostrils at 5-minute intervals. Sensations of nasal pain, blockage, and drip increased with concentration and were significantly elevated above normal. These parallels suggested activation of similar subsets of afferent neurons. Urea and lysozyme secretion were dose dependent in all groups, suggesting that serous cell exocytosis was one source of urea after neural stimulation. Only AR and normal groups had mucin dose responses and correlations between symptoms and lysozyme secretion (R2 = 0.12eC0.23). The lysozyme dose responses may represent axon responses in these groups. The neurogenic stimulus did not alter albumin (vascular) exudation in any group. Albumin and mucin concentrations were correlated in sinusitis, suggesting that nonneurogenic factors predominated in sinusitis mucous hypersecretion. CFS had neural hypersensitivity (pain) but reduced serous cell secretion. HTS nasal provocations identified significant, unique patterns of neural and mucosal dysregulation in each rhinosinusitis syndrome.
Key Words: axon response glandular exocytosis mucosal hyperresponsiveness neurogenic inflammation urea secretion
Mucosal hyperresponsiveness appears to be a property of inflamed mucosa and airways (1eC3). Distinct sets of molecular mechanisms may be implicated in allergic, infectious, and nonallergic disorders. These mechanisms may depend on different mediators, the distributions of their receptors on resident and inflammatory cells, and the consequences of stimulation (or inhibition) of these cells (2, 4). For example, histamine provocation stimulates superficial postcapillary venule endothelial cell contraction that permits hydrostatically driven vascular plasma exudation (1, 3, 5), stimulation of H1-receptoreCbearing type C "itch" neurons (6), bronchial smooth muscle contraction (2), and afferenteCefferent cholinergic reflex-mediated glandular exocytosis (4, 5). Methacholine stimulates muscarinic M3-receptors on glands and smooth muscle to produce exocytosis and contraction (2, 7, 8). Subjects with allergic rhinitis (AR) develop afferent responsiveness to bradykinin (9, 10) and endothelin-1 (11) with induction of afferenteCefferent reflexes. In humans, hypertonic saline (HTS) causes neural depolarization with local axon responses that release substance P in the mucosa (12, 13). Glandular secretion is stimulated with no alteration in plasma exudation (12). HTS-induced effects are accentuated in AR (14, 15). Togias and colleagues have shown that similar nociceptive neural responses may mediate the effects of cold dry air (16) because evaporation of water from the epithelial lining fluid can lead to local hypertonic conditions (17, 18). These findings suggest that mechanisms responsible for hyperresponsiveness may include increased sensitivity of afferent nociceptive neurons, induction of afferenteCefferent cholinergic parasympathetic reflexes, neural plasticity with inflammation-induced changes in the combinations of neuropeptides released by various subpopulations of neurons, the potential that different neural populations have distinct patterns of local mucosal axon response effects, as well as inflammation-induced changes in glandular secretion. Additional mechanisms are likely to co-exist.
HTS nasal provocations were used to compare neural responses (19, 20) between AR, acute sinusitis, and the nonallergic rhinitis present in subjects with chronic fatigue syndrome (CFS). Symptomatic differences between these groups and normal control subjects were assessed by rhinitis (19, 21, 22) and sinusitis (23) questionnaires and measurement of HTS-induced sensations by linear analog scales. Functional mucosal responses were assessed in lavage fluids collected from unilateral HTS-challenged and contralateral "unperturbed" nostrils. Marker proteins were albumin (plasma exudation), mucin (submucosal gland mucous and epithelial goblet cells), lysozyme (serous cells of glands), and urea, a presumed measure of interstitial fluid flux (5, 7, 9, 12, 24, 25). The slopes of these dose responses were compared between the pathophysiologically distinct conditions to assess differences in nociception (pain), other sensations, axon response effects, allergic and nonallergic inflammatory effects on vascular permeability and glandular secretion, and the relationships between these independent variables. Some results from this study have been presented as abstracts (23, 26).
METHODS
Subjects
All participants gave informed consent for this paid, institutional review boardeCapproved, parallel-group study. Active AR (27), acute sinusitis (28, 29), and CFS (20) were defined by standard criteria (see the online supplement). Subjects completed a new sinusitis questionnaire (23) and our established rhinitis score (21, 22). Normal control subjects had no viral or other rhinitis complaints in the previous 6 weeks. Subjects were excluded for any steroid treatment in the past month, chronic illnesses, or use of drugs that could potentially affect afferent neuron, gland secretion, or plasma leak (see the online supplement).
HTS Provocation Protocol
Pre-existing secretions were removed by lavaging each nostril with saline until the discharge was clear (4, 7, 9, 12, 24, 25). Saline provocations (100 e) (one times tonicity of normal saline, 0.9% NaCl) were sprayed into both nostrils. Five minutes later, subjects completed the symptoms questionnaires. The right (ipsilateral) and then the left (contralateral) nostrils were separately lavaged with saline (1 ml by Beconase applicator). Next, three times the tonicity of normal saline (2.7%; 100 e) was sprayed into the right nostril with 100 e one times the tonicity of normal saline to the left (12). Five minutes later, lavage and questionnaires were again completed. This cycle was repeated with 6 times the tonicity of normal saline (5.4%), 12 times the tonicity of normal saline (10.8%), and 24 times the tonicity of normal saline (21.6%) to the right nostril and 100 e one times the tonicity of normal saline on the left. Secretions were chilled, weighed, and frozen (eC20°C).
Subjective Responses
The initial spike of "first pain" (30) was scored as pain intensity on a 10-cm linear analog scale anchored by "no pain" (0 cm) and "most severe pain in my life" (10 cm). The pain duration of the dull, paresthetic "second pain" (30) was timed (minutes). The sensation of nasal blockage or difficulty breathing through the left and right nostrils was scored after 5 minutes as "none" to "totally blocked." The sensation of nasal drip from the anterior nostrils was scored as "none" to "most severe in my life."
Lavage Fluid Analysis
Aliquots (25 e) were assayed in triplicate for albumin, urea, mucous cell acidic mucins, and serous cell lysozyme (5, 7, 9, 12, 24, 25).
Statistical Analysis of HTS DoseeCResponse Curves
Responses to one times the tonicity of normal saline and three times the tonicity of normal saline did not differ within each group (paired Student's t tests), and thus, three times the tonicity of normal saline results became the initial point for measuring the slopes for every individual HTS dose response. The logarithms of HTS doses were used. The mean slopes for each group were first compared for differences from zero to define significant dose responses (31). Second, slopes were compared between groups by two-tailed, unpaired Student's t tests with Bonferroni corrections for multiple comparisons (32) to identify any potentially important pathophysiologic differences between groups. Third, the midpoint responses at the geometric mean (8.5 times the tonicity of normal saline) were calculated for each variable and compared between groups by corrected two-tailed unpaired Student's t tests. Significance was defined by a p of less than 0.05, and values were reported as means with 95% confidence intervals (31, 32).
Linear Regression for Explained Variances (R2) Between Variables
Scientific significance was assumed for R2 0.15 (p < 0.05 for Pearson's correlation coefficients) (31, 32).
RESULTS
Demographic and Questionnaire Data
The ages of the sinusitis (mean, 42.6 years; 95% confidence interval, 36.8 to 48.4; n = 25, p < 0.05) and CFS (46.6 years, 42.6 to 50.7, n = 14, p < 0.0001) groups were significantly greater than the normal group (29.1 years, 25.6 to 32.6, n = 20, multiple analyses of variance (MANOVA) for comparisons of means from all groups, followed by two-tailed unpaired Student's t tests with Bonferroni corrections for multiple comparisons for differences in means between pairs of groups). The sinusitis group was older than the AR group (31.6 years, 25.2 to 35.0, n = 17, p < 0.001). The percentage of females did not differ significantly between groups.
Rhinitis score questionnaires asked subjects to recall and assess symptoms over the previous week. Rhinitis scores were significantly higher for sinusitis, AR, and CFS groups than normal (Table 1). Chest scores were similar for sinusitis and CFS. AR, sinusitis, and CFS groups had higher chest scores than normal.
Sinusitis questionnaires assessed symptoms on the day of the provocation. Scores for all domains for the normal group were essentially zero and thus were significantly lower than for the other three groups (p < 0.0001) (Table 1). Sinusitis tended to have the highest scores, followed closely by scores for the CFS group. The sinusitis group had significantly higher sinus pain and fullness, pharynx, bother, and total sinusitis scores compared with AR. Subjects with CFS had higher pharynx scores than AR. This was not surprising as sore throat is a minor case designation criterion for CFS (22). The proportions of each group that exceeded the predefined threshold scores for each domain (see online supplement) were also significantly different from normal (2 tests) (Table 1). Differences between groups for each individual questionnaire item and the identification of any potentially disease-specific items will be ascertained in future analyses.
HTS Provocation Dose Responses
Pain intensity.
Pain intensities increased in HTS doseeCdependent fashion for each group (Figure 1). Slopes were significantly greater than zero (p < 0.001). Only the slope for CFS was steeper than the normal response (p < 0.05). The CFS and sinusitis dose responses were shifted to the left of normal (p < 0.05). The pain intensities at the geometric mean midpoints were highest for CFS and sinusitis compared with normal (p < 0.05) and AR (p < 0.05) (Table 2). The subjects with CFS had more intense pain sensations than the normal group.
Pain duration.
HTS dose responses for all groups were significantly different from zero but were not different from each other (see Figure E1 in the online supplement). There was a nonsignificant trend for duration to be longer in CFS (2.41 minutes) and sinusitis (2.32 minutes) than AR (1.72 minutes) and normal (1.86 minutes) (Table 2).
Sensation of nasal blockage.
The sinusitis, AR, and CFS doseeCresponse slopes were not different from each other (Figure 2). The slope for sinusitis was the flattest but was still significantly different from zero (p < 0.05). The midpoints for the sinusitis, AR, and CFS doseeCresponse curves were all shifted significantly upward compared with normal (p < 0.001) (Table 2).
Sensation of drip/perception of rhinorrhea.
Drip mirrored blockage. Slopes were greater than zero for all four curves (p < 0.001) and did not differ from each other (Figure E2). However, the sinusitis (p < 0.001), AR (p < 0.01), and CFS (p < 0.001) dose responses were significantly shifted upward compared with normal (Table 2).
Weights of returned lavage fluids.
There was a trend for HTS to stimulate higher secretion weights from the AR and normal compared with CFS and sinusitis groups (Figure E3). The AR doseeCresponse curve was shifted upward from the sinusitis response (p < 0.05). No other comparisons of slopes or midpoints were significant.
Mucin.
Mucin was significantly lower for the normal group at three times the tonicity of normal saline than each of the other groups (Figure 3). In general, mucin concentrations were the highest in the sinusitis group. However, the wide variances for mucin measurements prevented significant differences between groups (Table 2). These variations in mucus production may be indicative of glandular dysregulation in sinusitis and AR. DoseeCresponse slopes were significantly greater than zero for AR (p < 0.05) and normal (p < 0.01) but were flat (p > 0.10) for sinusitis and CFS. HTS did not significantly alter mucin exocytosis and secretion in the CFS group.
Lysozyme.
DoseeCresponse slopes were significantly greater than zero for sinusitis (p < 0.001), normal (p < 0.001), AR (p < 0.05), and CFS (p < 0.05) (Figure E4). Variances were large for each measurement, but significant trends were identified for groups. The midpoint was significantly higher for AR than CFS (p < 0.01) and normal (p < 0.05) groups (Table 2). Sinusitis was also higher than CFS (p < 0.05).
Albumin.
Albumin concentrations in lavage fluid did not change with HTS provocation in any group (Figure 4). None of the slopes differed from zero. The sinusitis and AR groups had similar albumin levels (Table 2) that were significantly higher than CFS (p < 0.01 and 0.05, respectively) and normal (p < 0.001 and 0.01, respectively) groups. More albumin was present in the CFS group than the normal group (p < 0.05; Table 2). This suggested that the mechanisms responsible for sinusitis, AR, and the non-AR of CFS caused excessive, continuous albumin exudation.
Urea.
HTS caused significant dose responses for all groups (Figure 5). The four doseeCresponse curves overlapped. AR (highest concentration) and CFS (lowest) were different at the p = 0.06 level (Table 2).
Contralateral Effects
There were no subjective or objective changes on the opposite nostril that received serial doses of one times the tonicity of normal saline only. This indicated that the intensity of these HTS provocations was insufficient to recruit parasympathetic reflexes in any of the four conditions.
Explained Variances Between HTS-induced Symptoms and Analytes for Each Group
Normal subjects showed significant (p < 0.05) relationships between drip and block (R2 = 0.16) and drip and weight (R2 = 0.16). Lysozyme was associated with weight (R2 = 0.20), urea (R2 = 0.21), and mucin (R2 = 0.29). However, mucin and urea were not related (R2 = 0.07). This suggested that nasal sensations of drip and blockage were correlated with markers of glandular exocytosis after this neurogenic stimulus.
Subjects with CFS had a relationship between lysozyme and urea (R2 = 0.29). However, lysozyme and mucin (R2 = 0.07) and mucin and urea (R2 = 0.07) were not related. This suggested that HTS induced parallel secretion of urea and lysozyme from serous cells but that there was dysfunction of lysozyme and mucin (serous and mucous cell) exocytosis. Weight was related to albumin (R2 = 0.17). Although drip and block were related to each other (R2 = 0.18), they were not related to any of the analytes. As a result, neural perceptions (sensations) were dissociated from the mucosal responses to HTS.
Sinusitis data revealed a strong relationship between albumin and mucin (R2 = 0.47). Albumin (R2 = 0.29) and mucin (R2 = 0.15) were associated with block, suggesting that nasal obstruction may have been related to vascular engorgement, plasma exudation, and mucous hypersecretion. Drip was associated with block (R2 = 0.21), weight (R2 = 0.18), and lysozyme (R2 = 0.23) but not other variables.
AR symptoms were weakly related to each other. Weight was related to mucin (R2 = 0.28), urea (R2 = 0.25), and lysozyme (R2 = 0.18). The association of lysozyme and mucin (R2 = 0.24) reinforced the important role of glandular secretion in AR symptomatology. Albumin and drip (R2 = 0.13) were of clinical relevance because watery rhinorrhea is a cardinal sign of acute allergen exposure. Urea was not related to any of the other analytes.
DISCUSSION
HTS-induced pain intensity curves for sinusitis and CFS groups were shifted to the left of the normal group. This may indicate greater "first pain" A nerve fiber sensitivity to hypertonicity (30), decreased spinal cord dorsal horn inhibition of pain, other differences in visceral (mucosal) sensing or perception (33), or greater CNS neural activation (34). Nasal block was much higher at all HTS doses for sinusitis, AR, and CFS compared with normal subjects. Drip dose responses in sinusitis, AR, and CFS were identical and elevated and shifted to the left compared with the normal curve. These parallel data suggested that afferent neurologic pathways or central perceptions of visceral, nasal sensations ("interoception") (33) may have been similarly affected by a single inflammatory or neuromodulatory mechanism in these three illness groups.
Drip, block, and glandular secretion were significantly associated in normal subjects. This suggested that HTS induced pain, nasal discharge, and these two sensations. This is consistent with our previous findings (12, 19, 23, 26, 35) and the hypothesis that HTS stimulated type C neurons that convey these sensations and have axon responses that stimulate glandular exocytosis. Results such as these indicate that subjective descriptors such as drip, block, "fullness," "congestion," and objective measures such as nasal airflow resistance and nasal airspace volume measured by acoustic rhinometry should not be used synonymously. Instead, the distinct mechanisms leading to conscious, subjective sensations must be distinguished from mucosal pathologic events such as plasma extravasation and watery discharge, glandular exocytosis of mucinous gels, and mucosal thickening caused by engorgement of venous sinusoids (4). For example, it has been shown that HTS stimulated equal magnitudes of sensations and secretions before and after nasal decongestion with oxymetazoline (35). HTS induced the same sensations of nasal blockage even when the nasal vessels were maximally constricted and the nasal cavity maximally patent (measured by acoustic rhinometry). This suggested that the sensation of blockage could be appreciated despite having no change in nasal patency. This could explain the paradoxical finding of subjective nasal obstruction in subjects without AR with patent, nonobstructed nostrils.
HTS had no effect on albumin exudation (12, 35). However, capsaicin doses that cause pain ratings of 7 to 10 out of 10 can induce vascular leak in subjects with untreated AR (13, 14, 36). These pain levels are not typically encountered in daily life or AR (itch predominates) (6). It is possible that other non-neurogenic mechanisms may regulate plasma extravasation. For example, denervated nasal polyps have a 10-fold higher plasma flux than normal nasal mucosa (37).
Previous studies have implicated neural axon responses in HTS nasal provocations (12). HTS dose responses cause incremental increases in nerve stimulation and sensations of pain, block, and drip (12). Substance P was released in a dose-dependent manner in the first 3 minutes of HTS provocations. Substance P containing neurons innervate glands (4). These glands express substance PeCpreferring neurokinin-1 receptors (12). The maximum glandular secretion and sensations of block and drip occurred between 3 and 5 minutes. Volume changes detected by acoustic rhinometry between the administration of each HTS dose and 5 minutes later were roughly equivalent to the net mass of secretions blown out of the nostrils (35). This sequence of neural stimulation, pain, and substance P release followed by glandular secretion and sensations of nasal obstruction and rhinorrhea may define the nociceptive nerve axon response to hypertonic conditions in human airways. This response may be an instantly recruited mucosal defense mechanism to protect against the effects of inhaled irritants such as smoke (4). Evidence for axon responses in AR may have been detected as the HTS dose-dependent secretion of lysozyme.
Subjects with sinusitis had high levels of subjective complaints but the lowest recovered weight of lavage fluid. The lavage fluid may have become trapped behind swollen nasal structures, rapidly absorbed into the epithelial lining fluid or tissue, and become unavailable for collection. The relationships between lysozyme, drip, block, and weight demonstrated the importance of glandular secretion for symptom production in sinusitis. Albumin, mucin, and block were related, suggesting that some aspect of sinusitis mucosal inflammation led to increased plasma extravasation, possible venous sinusoid engorgement, and goblet and glandular mucous hypersecretion.
CFS was anomalous. Sinus pain and fullness responses were elevated and not significantly different from sinusitis (Table 2). We have previously shown that subjects with CFS have lower pain thresholds (increased tenderness to pressure) over their maxillary, frontal, and ethmoid regions compared with normal subjects or subjects with AR, acute sinusitis, and chronic sinusitis (38). Subjects with CFS and sinusitis could be differentiated by low systemic pain thresholds in CFS but not sinusitis. Lower pain thresholds and potentially lower thresholds for other nociceptive stimuli such as HTS would be consistent with systemic nociceptive dysfunction in CFS. This concept was supported by the HTS dose responses for pain intensity, duration, block and drip. The magnitudes for these sensations were as severe in CFS as in sinusitis and AR. The cause of this neural hyperresponsiveness in CFS remains unclear but was unlikely to be related to the inflammatory mechanisms of sinusitis or AR (24, 39, 40). Mucin and albumin secretion were elevated compared with normal (41) but did not respond to HTS stimulation. Lysozyme and mucin dose responses were not correlated, suggesting glandular dysfunction with decreased serous cell but increased mucous cell exocytosis. The mechanism was not apparent.
The urea results were initially perplexing. Urea has been thought to passively diffuse from the interstitial fluid into the nasal epithelial lining fluid (25). Albumin has a similar pathway for secretion (5). However, urea increased in HTS dose-dependent fashion, whereas albumin did not. This suggested that the vascular permeability and transepithelial exudation, which account for the increased albumin exudation in conditions such as AR, did not play a role in urea transport after neurogenic stimulation. Instead, urea was correlated to lysozyme under these conditions. This implied that serous cells were the primary source of urea when these glands were stimulated by the inflammatory milieu, as in sinusitis, or by HTS provocation. The link between stimulation of pain-conveying neurons and glandular secretion may have been the nociceptive nerve axon response mechanism (12). Serous cell transport of urea may be analogous to secretion of uric acid, the major antioxidant in nasal secretions (42). Specific urea transport mechanisms in this mucosa have not been defined.
This pilot project had several limitations that can be rectified in future investigations. Unequal numbers of subjects were present in each group. However, highly significant differences in the magnitudes of changes from normal and between groups were still recognized after Bonferroni corrections for multiple comparisons. The subjects with AR had moderate to severe acute seasonal symptoms on the days of their provocations. Some also had perennial AR, but this did not appear to alter the nasal response. It was not clear whether the duration of symptoms affected their responses. Longitudinal studies beginning before a short allergen season in monosensitized subjects (e.g., birch pollen) may be needed to identify any progression of changes in neural sensitivities, central perceptions of symptom severity, and axon response mechanisms. As noted previously here, other provocation agents may have had different effects in these syndromes. It was not feasible to design a study that could simultaneously compare responses to methacholine, histamine, or other agents in tandem with HTS in the same subjects. Instead, additional parallel group studies would be required.
This preliminary study demonstrated differences in neurogenic responses for sinusitis, AR, and CFS from control subjects. Each inflammatory state had elevated HTS-induced sensations compared with normal but had distinct patterns of secreted proteins. This HTS model offers a reductionist approach to understand the neurogenic and other mechanisms that contribute to gland, vascular, and neural hyperresponsiveness in airway disorders.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Meyer P, Andersson M, Persson CG, Greiff L. Steroid-sensitive indices of airway inflammation in children with seasonal allergic rhinitis. Pediatr Allergy Immunol 2003;14:60eC65.
Dahl R, Mygind N. Mechanisms of airflow limitation in the nose and lungs. Clin Exp Allergy 1998;28:17eC25.
Svensson C, Andersson M, Greiff L, Persson CG. Nasal mucosal end organ hyperresponsiveness. Am J Rhinol 1998;12:1:37eC43.
Tai C-F, Baraniuk JN. Upper airway neurogenic mechanisms. Curr Opin Allergy Clin Immunol 2002;2:11eC19.
Raphael GD, Meredith SD, Baraniuk JN, Druce HM, Banks SM, Kaliner MA. The pathophysiology of rhinitis: II: assessment of the sources of protein in histamine-induced nasal secretions. Am Rev Respir Dis 1989;139:791eC800.
Nicolson TA, Bevan S, Richard CD. Characterisation of the calcium response to histamine in capsaicin-sensitive and capsaicin-insensitive sensory nerve. Neuroscience 2002;110:329eC338.
Raphael GD, Druce HM, Baraniuk JN, Kaliner MA. Pathophysiology of rhinitis: I: assessment of the sources of protein in methacholine-induced nasal secretions. Am Rev Respir Dis 1988;138:413eC420.
Mullol J, Baraniuk JN, Logun C, Merida M, Hausfeld JH, Shelhamer JH, Kaliner MA. M1 and M3 muscarinic antagonists inhibit human nasal glandular secretion in vitro. J Appl Physiol 1992;73:2069eC2073.
Baraniuk JN, Silver PB, Kaliner MA, Barnes PJ. Perennial rhinitis subjects have altered vascular, glandular, and neural responses to bradykinin nasal provocation. Int Arch Allergy Immunol 1994;103:202eC208.
Riccio MM, Proud D. Evidence that enhanced nasal reactivity to bradykinin in patients with symptomatic allergy is mediated by neural reflexes. J Allergy Clin Immunol 1996;97:1252eC1263.
Riccio MM, Reynolds CJ, Hay DW, Proud D. Effects of intranasal administration of endothelin-1 to allergic and nonallergic individuals. Am J Respir Crit Care Med 1995;152:1757eC1764.
Baraniuk JN, Ali M, Yuta A, Fang S-Y, Naranch K. Hypertonic saline nasal provocation stimulates nociceptive nerves, substance P release, and glandular mucus exocytosis in normal humans. Am J Respir Crit Care Med 1999;160:655eC662.
Sanico AM, Philip G, Lai GK, Togias A. Hyperosmolar saline induces reflex nasal secretions, evincing neural hyperresponsiveness in allergic rhinitis. J Appl Physiol 1999;86:1202eC1210.
Sanico AM, Stanisz AM, Gleeson TD, Bora S, Proud D, Bienenstock J, Koliatsos VE, Togias A. Nerve growth factor expression and release in allergic inflammatory disease of the upper airways. Am J Respir Crit Care Med 2000;161:1631eC1635.
Koskela H, Di Sciascio MB, Anderson SD, Andersson M, Chan HK, Gadalla S, Katelaris C. Nasal hyperosmolar challenge with a dry powder of mannitol in patients with allergic rhinitis: evidence for epithelial cell involvement. Clin Exp Allergy 2000;30:1627eC1636.
Togias A, Lykens K, Kagey-Sobotka A, Eggleston PA, Proud D, Lichtenstein LM, Naclerio RM. Studies on the relationships between sensitivity to cold, dry air, hyperosmolar solutions and histamine in the adult nose. Am Rev Respir Dis 1990;141:1428eC1433.
Anderson SD, Spring J, Moore B, Rodwell LT, Spalding N, Gonda I, Chan K, Walsh A, Clark AR. The effect of inhaling a dry powder of sodium chloride on the airways of asthmatic subjects. Eur Respir J 1997;10:2465eC2473.
Togias AG, Proud D, Lichtenstein LM, Adams GK III, Norman PS, Kagey-Sobotka A, Naclerio RM. The osmolality of nasal secretions increases when inflammatory mediators are released in response to inhalation of cold, dry air. Am Rev Respir Dis 1988;137:625eC629.
Baraniuk JN, Clauw DJ, Gaumond E. Rhinitis symptoms in chronic fatigue syndrome. Ann Allergy Asthma Immunol 1998;81:359eC365.
Fukuda K, Straus SE, Hickei I, Sharpe MC, Dobbins JC, Komaroff A. The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med 1994;121:953eC959.
Wasserfallen JB, Gold K, Schulman KA, Milzman D, Baraniuk JN. Development and validation of a rhinoconjunctivitis and asthma symptom score for use as an outcome measure in clinical trials. J Allergy Clin Immunol 1997;100:16eC22.
Wasserfallen JB, Gold K, Schulman KA, Baraniuk JN. Item responsiveness of a rhinitis and asthma symptoms score during a pollen season. J Asthma 1999;36:459eC465.
Baraniuk JN, Yuta A, Tai C-F, Park Y-J. Neurogenic responses in rhinosinusitis . Am J Respir Crit Care Med 2004;169:A820.
Baraniuk JN, Clauw DJ, Yuta A, Ali M, Gaumond E, Upadhyayula N, Fujita K, Shimizu T. Nasal secretion analysis in allergic rhinitis, cystic fibrosis and fibromyalgia / chronic fatigue syndrome subjects. Am J Rhinol 1998;12:435eC440.
Kaulbach HC, White MV, Igarashi Y, Hahn BK, Kaliner MA. Estimation of nasal epithelial lining fluid using urea as a marker. J Allergy Clin Immunol 1993;92:457eC465.
Baraniuk JN, Naranch K, Ali M, Le U. Pathophysiology of rhinosinusitis: baseline symptoms and secretions in acute sinusitis (S), allergic rhinitis (AR), the nonallergic rhinitis of chronic fatigue syndrome (CFS) and normal subjects . Am J Respir Crit Care Med 2004;169:A820.
van Cauwenberge PB, Wang D-Y, Ingels KJAO, Bachert C. Rhinitis: the spectrum of disease. In: Busse WW, Holgate ST, editors. Asthma and rhinitis, 2nd ed. Oxford: Blackwell Science; 2000. p. 6eC16.
Kaliner MA, Osguthorpe JD, Fireman P, Anon J, Georgitis J, Davis ML, Naclerio R, Kennedy D. Sinusitis: bench to bedside: current findings, future directions. Otolaryngol Head Neck Surg 1997;116:S1eCS20.
Benninger MS, Ferguson BJ, Hadley JA, Hamilos DL, Jacobs M, Kennedy DW, Lanza DC, Marple BF, Osguthorpe JD, Stankiewicz JA, et al. Adult chronic rhinosinusitis: definitions, diagnosis, epidemiology, and pathophysiology. Otolaryngol Head Neck Surg 2003;129:S1eCS32.
Dray A, Urban L, Dickenson A. Pharmacology of chronic pain. Trends Pharmacol Sci 1994;15:190eC197.
Swinscow TDV, Campbell MJ. Statistics at square one. London: BMJ Books; 2002. p. 44eC51.
Gardner MJ, Altman DG, editors. Statistics with confidence: confidence intervals and statistical guidelines. London: BMJ Books; 1989. pp. 1eC140.
Adam G. Visceral perception: understanding internal organs. New York: Plenum Press; 1998.
Coghill RC, McHaffie JC, Yen Y-F. Neural correlates of interindividual differences in the subjective experience of pain. Proc Natl Acad Sci USA 2003;100:8538eC8542.
Baraniuk JN, Ali M, Naranch K. Hypertonic saline nasal provocation and acoustic rhinometry. Clin Exp Allergy 2002;32:543eC550.
Sanico AM, Atsuta S, Proud D, Togias A. Plasma extravasation through neuronal stimulation in human nasal mucosa in the setting of allergic rhinitis. J Appl Physiol 1998;84:537eC543.
Biewenga J, Stoop AE, van der Heijden HA, van der Baan S, van Kamp GJ. Albumin and immunoglobulin levels in nasal secretions of patients with nasal polyps treated with endoscopic sinus surgery and topical corticosteroids. J Allergy Clin Immunol 1995;96:334eC340.
Naranch K, Park Y-J, Repka-Ramirez SM, Velarde A, Clauw D, Baraniuk JN. A tender sinus does not always mean sinusitis. Otolaryngol Head Neck Surg 2002;127:387eC397.
Repka-Ramirez MS, Naranch K, Park Y-J, Clauw D, Baraniuk JN. Cytokines in nasal lavage fluids from acute sinusitis, allergic rhinitis, and chronic fatigue syndrome subjects. Allergy Asthma Proc 2002;23:185eC190.
Repka-Ramirez MS, Naranch K, Park Y-J, Velarde A, Clauw D, Baraniuk JN. IgE levels are the same in chronic fatigue syndrome (CFS) and control subjects when stratified by allergy skin test results and rhinitis types. Ann Allergy Asthma Immunol 2001;87:218eC221.
Naranch K, Repka-Ramirez SM, Park Y-J, Velarde A, Finnegan R, Murray J, Pheiffer A, Hwang E, Clauw D, Baraniuk JN. Differences in baseline nasal secretions between chronic fatigue syndrome (CFS) and control subjects. J Chronic Fatigue Syndr 2002;10:3eC15.
Peden DB, Hohman R, Brown ME, Mason RT, Berkebile C, Fales HM, Kaliner MA. Uric acid is a major antioxidant in human nasal airway secretions. Proc Natl Acad Sci USA 1990;87:7638eC7642.(James N. Baraniuk, Kristi)
Department of Otorhinolaryngology, Kaohsiung Medical University, Kaohsiung City, Taiwan
Department of Otorhinolaryngology, Catholic University, Seoul, Korea
Department of Otorhinolaryngology, Mie University, Tsu, Japan
ABSTRACT
Pathophysiologic differences in neural responses to hypertonic saline (HTS) were investigated in subjects with acute sinusitis (n = 25), subjects with chronic fatigue syndrome (CFS) with nonallergic rhinitis (n = 14), subjects with active allergic rhinitis (AR; n = 17), and normal (n = 20) subjects. Increasing strengths of HTS were sprayed into their nostrils at 5-minute intervals. Sensations of nasal pain, blockage, and drip increased with concentration and were significantly elevated above normal. These parallels suggested activation of similar subsets of afferent neurons. Urea and lysozyme secretion were dose dependent in all groups, suggesting that serous cell exocytosis was one source of urea after neural stimulation. Only AR and normal groups had mucin dose responses and correlations between symptoms and lysozyme secretion (R2 = 0.12eC0.23). The lysozyme dose responses may represent axon responses in these groups. The neurogenic stimulus did not alter albumin (vascular) exudation in any group. Albumin and mucin concentrations were correlated in sinusitis, suggesting that nonneurogenic factors predominated in sinusitis mucous hypersecretion. CFS had neural hypersensitivity (pain) but reduced serous cell secretion. HTS nasal provocations identified significant, unique patterns of neural and mucosal dysregulation in each rhinosinusitis syndrome.
Key Words: axon response glandular exocytosis mucosal hyperresponsiveness neurogenic inflammation urea secretion
Mucosal hyperresponsiveness appears to be a property of inflamed mucosa and airways (1eC3). Distinct sets of molecular mechanisms may be implicated in allergic, infectious, and nonallergic disorders. These mechanisms may depend on different mediators, the distributions of their receptors on resident and inflammatory cells, and the consequences of stimulation (or inhibition) of these cells (2, 4). For example, histamine provocation stimulates superficial postcapillary venule endothelial cell contraction that permits hydrostatically driven vascular plasma exudation (1, 3, 5), stimulation of H1-receptoreCbearing type C "itch" neurons (6), bronchial smooth muscle contraction (2), and afferenteCefferent cholinergic reflex-mediated glandular exocytosis (4, 5). Methacholine stimulates muscarinic M3-receptors on glands and smooth muscle to produce exocytosis and contraction (2, 7, 8). Subjects with allergic rhinitis (AR) develop afferent responsiveness to bradykinin (9, 10) and endothelin-1 (11) with induction of afferenteCefferent reflexes. In humans, hypertonic saline (HTS) causes neural depolarization with local axon responses that release substance P in the mucosa (12, 13). Glandular secretion is stimulated with no alteration in plasma exudation (12). HTS-induced effects are accentuated in AR (14, 15). Togias and colleagues have shown that similar nociceptive neural responses may mediate the effects of cold dry air (16) because evaporation of water from the epithelial lining fluid can lead to local hypertonic conditions (17, 18). These findings suggest that mechanisms responsible for hyperresponsiveness may include increased sensitivity of afferent nociceptive neurons, induction of afferenteCefferent cholinergic parasympathetic reflexes, neural plasticity with inflammation-induced changes in the combinations of neuropeptides released by various subpopulations of neurons, the potential that different neural populations have distinct patterns of local mucosal axon response effects, as well as inflammation-induced changes in glandular secretion. Additional mechanisms are likely to co-exist.
HTS nasal provocations were used to compare neural responses (19, 20) between AR, acute sinusitis, and the nonallergic rhinitis present in subjects with chronic fatigue syndrome (CFS). Symptomatic differences between these groups and normal control subjects were assessed by rhinitis (19, 21, 22) and sinusitis (23) questionnaires and measurement of HTS-induced sensations by linear analog scales. Functional mucosal responses were assessed in lavage fluids collected from unilateral HTS-challenged and contralateral "unperturbed" nostrils. Marker proteins were albumin (plasma exudation), mucin (submucosal gland mucous and epithelial goblet cells), lysozyme (serous cells of glands), and urea, a presumed measure of interstitial fluid flux (5, 7, 9, 12, 24, 25). The slopes of these dose responses were compared between the pathophysiologically distinct conditions to assess differences in nociception (pain), other sensations, axon response effects, allergic and nonallergic inflammatory effects on vascular permeability and glandular secretion, and the relationships between these independent variables. Some results from this study have been presented as abstracts (23, 26).
METHODS
Subjects
All participants gave informed consent for this paid, institutional review boardeCapproved, parallel-group study. Active AR (27), acute sinusitis (28, 29), and CFS (20) were defined by standard criteria (see the online supplement). Subjects completed a new sinusitis questionnaire (23) and our established rhinitis score (21, 22). Normal control subjects had no viral or other rhinitis complaints in the previous 6 weeks. Subjects were excluded for any steroid treatment in the past month, chronic illnesses, or use of drugs that could potentially affect afferent neuron, gland secretion, or plasma leak (see the online supplement).
HTS Provocation Protocol
Pre-existing secretions were removed by lavaging each nostril with saline until the discharge was clear (4, 7, 9, 12, 24, 25). Saline provocations (100 e) (one times tonicity of normal saline, 0.9% NaCl) were sprayed into both nostrils. Five minutes later, subjects completed the symptoms questionnaires. The right (ipsilateral) and then the left (contralateral) nostrils were separately lavaged with saline (1 ml by Beconase applicator). Next, three times the tonicity of normal saline (2.7%; 100 e) was sprayed into the right nostril with 100 e one times the tonicity of normal saline to the left (12). Five minutes later, lavage and questionnaires were again completed. This cycle was repeated with 6 times the tonicity of normal saline (5.4%), 12 times the tonicity of normal saline (10.8%), and 24 times the tonicity of normal saline (21.6%) to the right nostril and 100 e one times the tonicity of normal saline on the left. Secretions were chilled, weighed, and frozen (eC20°C).
Subjective Responses
The initial spike of "first pain" (30) was scored as pain intensity on a 10-cm linear analog scale anchored by "no pain" (0 cm) and "most severe pain in my life" (10 cm). The pain duration of the dull, paresthetic "second pain" (30) was timed (minutes). The sensation of nasal blockage or difficulty breathing through the left and right nostrils was scored after 5 minutes as "none" to "totally blocked." The sensation of nasal drip from the anterior nostrils was scored as "none" to "most severe in my life."
Lavage Fluid Analysis
Aliquots (25 e) were assayed in triplicate for albumin, urea, mucous cell acidic mucins, and serous cell lysozyme (5, 7, 9, 12, 24, 25).
Statistical Analysis of HTS DoseeCResponse Curves
Responses to one times the tonicity of normal saline and three times the tonicity of normal saline did not differ within each group (paired Student's t tests), and thus, three times the tonicity of normal saline results became the initial point for measuring the slopes for every individual HTS dose response. The logarithms of HTS doses were used. The mean slopes for each group were first compared for differences from zero to define significant dose responses (31). Second, slopes were compared between groups by two-tailed, unpaired Student's t tests with Bonferroni corrections for multiple comparisons (32) to identify any potentially important pathophysiologic differences between groups. Third, the midpoint responses at the geometric mean (8.5 times the tonicity of normal saline) were calculated for each variable and compared between groups by corrected two-tailed unpaired Student's t tests. Significance was defined by a p of less than 0.05, and values were reported as means with 95% confidence intervals (31, 32).
Linear Regression for Explained Variances (R2) Between Variables
Scientific significance was assumed for R2 0.15 (p < 0.05 for Pearson's correlation coefficients) (31, 32).
RESULTS
Demographic and Questionnaire Data
The ages of the sinusitis (mean, 42.6 years; 95% confidence interval, 36.8 to 48.4; n = 25, p < 0.05) and CFS (46.6 years, 42.6 to 50.7, n = 14, p < 0.0001) groups were significantly greater than the normal group (29.1 years, 25.6 to 32.6, n = 20, multiple analyses of variance (MANOVA) for comparisons of means from all groups, followed by two-tailed unpaired Student's t tests with Bonferroni corrections for multiple comparisons for differences in means between pairs of groups). The sinusitis group was older than the AR group (31.6 years, 25.2 to 35.0, n = 17, p < 0.001). The percentage of females did not differ significantly between groups.
Rhinitis score questionnaires asked subjects to recall and assess symptoms over the previous week. Rhinitis scores were significantly higher for sinusitis, AR, and CFS groups than normal (Table 1). Chest scores were similar for sinusitis and CFS. AR, sinusitis, and CFS groups had higher chest scores than normal.
Sinusitis questionnaires assessed symptoms on the day of the provocation. Scores for all domains for the normal group were essentially zero and thus were significantly lower than for the other three groups (p < 0.0001) (Table 1). Sinusitis tended to have the highest scores, followed closely by scores for the CFS group. The sinusitis group had significantly higher sinus pain and fullness, pharynx, bother, and total sinusitis scores compared with AR. Subjects with CFS had higher pharynx scores than AR. This was not surprising as sore throat is a minor case designation criterion for CFS (22). The proportions of each group that exceeded the predefined threshold scores for each domain (see online supplement) were also significantly different from normal (2 tests) (Table 1). Differences between groups for each individual questionnaire item and the identification of any potentially disease-specific items will be ascertained in future analyses.
HTS Provocation Dose Responses
Pain intensity.
Pain intensities increased in HTS doseeCdependent fashion for each group (Figure 1). Slopes were significantly greater than zero (p < 0.001). Only the slope for CFS was steeper than the normal response (p < 0.05). The CFS and sinusitis dose responses were shifted to the left of normal (p < 0.05). The pain intensities at the geometric mean midpoints were highest for CFS and sinusitis compared with normal (p < 0.05) and AR (p < 0.05) (Table 2). The subjects with CFS had more intense pain sensations than the normal group.
Pain duration.
HTS dose responses for all groups were significantly different from zero but were not different from each other (see Figure E1 in the online supplement). There was a nonsignificant trend for duration to be longer in CFS (2.41 minutes) and sinusitis (2.32 minutes) than AR (1.72 minutes) and normal (1.86 minutes) (Table 2).
Sensation of nasal blockage.
The sinusitis, AR, and CFS doseeCresponse slopes were not different from each other (Figure 2). The slope for sinusitis was the flattest but was still significantly different from zero (p < 0.05). The midpoints for the sinusitis, AR, and CFS doseeCresponse curves were all shifted significantly upward compared with normal (p < 0.001) (Table 2).
Sensation of drip/perception of rhinorrhea.
Drip mirrored blockage. Slopes were greater than zero for all four curves (p < 0.001) and did not differ from each other (Figure E2). However, the sinusitis (p < 0.001), AR (p < 0.01), and CFS (p < 0.001) dose responses were significantly shifted upward compared with normal (Table 2).
Weights of returned lavage fluids.
There was a trend for HTS to stimulate higher secretion weights from the AR and normal compared with CFS and sinusitis groups (Figure E3). The AR doseeCresponse curve was shifted upward from the sinusitis response (p < 0.05). No other comparisons of slopes or midpoints were significant.
Mucin.
Mucin was significantly lower for the normal group at three times the tonicity of normal saline than each of the other groups (Figure 3). In general, mucin concentrations were the highest in the sinusitis group. However, the wide variances for mucin measurements prevented significant differences between groups (Table 2). These variations in mucus production may be indicative of glandular dysregulation in sinusitis and AR. DoseeCresponse slopes were significantly greater than zero for AR (p < 0.05) and normal (p < 0.01) but were flat (p > 0.10) for sinusitis and CFS. HTS did not significantly alter mucin exocytosis and secretion in the CFS group.
Lysozyme.
DoseeCresponse slopes were significantly greater than zero for sinusitis (p < 0.001), normal (p < 0.001), AR (p < 0.05), and CFS (p < 0.05) (Figure E4). Variances were large for each measurement, but significant trends were identified for groups. The midpoint was significantly higher for AR than CFS (p < 0.01) and normal (p < 0.05) groups (Table 2). Sinusitis was also higher than CFS (p < 0.05).
Albumin.
Albumin concentrations in lavage fluid did not change with HTS provocation in any group (Figure 4). None of the slopes differed from zero. The sinusitis and AR groups had similar albumin levels (Table 2) that were significantly higher than CFS (p < 0.01 and 0.05, respectively) and normal (p < 0.001 and 0.01, respectively) groups. More albumin was present in the CFS group than the normal group (p < 0.05; Table 2). This suggested that the mechanisms responsible for sinusitis, AR, and the non-AR of CFS caused excessive, continuous albumin exudation.
Urea.
HTS caused significant dose responses for all groups (Figure 5). The four doseeCresponse curves overlapped. AR (highest concentration) and CFS (lowest) were different at the p = 0.06 level (Table 2).
Contralateral Effects
There were no subjective or objective changes on the opposite nostril that received serial doses of one times the tonicity of normal saline only. This indicated that the intensity of these HTS provocations was insufficient to recruit parasympathetic reflexes in any of the four conditions.
Explained Variances Between HTS-induced Symptoms and Analytes for Each Group
Normal subjects showed significant (p < 0.05) relationships between drip and block (R2 = 0.16) and drip and weight (R2 = 0.16). Lysozyme was associated with weight (R2 = 0.20), urea (R2 = 0.21), and mucin (R2 = 0.29). However, mucin and urea were not related (R2 = 0.07). This suggested that nasal sensations of drip and blockage were correlated with markers of glandular exocytosis after this neurogenic stimulus.
Subjects with CFS had a relationship between lysozyme and urea (R2 = 0.29). However, lysozyme and mucin (R2 = 0.07) and mucin and urea (R2 = 0.07) were not related. This suggested that HTS induced parallel secretion of urea and lysozyme from serous cells but that there was dysfunction of lysozyme and mucin (serous and mucous cell) exocytosis. Weight was related to albumin (R2 = 0.17). Although drip and block were related to each other (R2 = 0.18), they were not related to any of the analytes. As a result, neural perceptions (sensations) were dissociated from the mucosal responses to HTS.
Sinusitis data revealed a strong relationship between albumin and mucin (R2 = 0.47). Albumin (R2 = 0.29) and mucin (R2 = 0.15) were associated with block, suggesting that nasal obstruction may have been related to vascular engorgement, plasma exudation, and mucous hypersecretion. Drip was associated with block (R2 = 0.21), weight (R2 = 0.18), and lysozyme (R2 = 0.23) but not other variables.
AR symptoms were weakly related to each other. Weight was related to mucin (R2 = 0.28), urea (R2 = 0.25), and lysozyme (R2 = 0.18). The association of lysozyme and mucin (R2 = 0.24) reinforced the important role of glandular secretion in AR symptomatology. Albumin and drip (R2 = 0.13) were of clinical relevance because watery rhinorrhea is a cardinal sign of acute allergen exposure. Urea was not related to any of the other analytes.
DISCUSSION
HTS-induced pain intensity curves for sinusitis and CFS groups were shifted to the left of the normal group. This may indicate greater "first pain" A nerve fiber sensitivity to hypertonicity (30), decreased spinal cord dorsal horn inhibition of pain, other differences in visceral (mucosal) sensing or perception (33), or greater CNS neural activation (34). Nasal block was much higher at all HTS doses for sinusitis, AR, and CFS compared with normal subjects. Drip dose responses in sinusitis, AR, and CFS were identical and elevated and shifted to the left compared with the normal curve. These parallel data suggested that afferent neurologic pathways or central perceptions of visceral, nasal sensations ("interoception") (33) may have been similarly affected by a single inflammatory or neuromodulatory mechanism in these three illness groups.
Drip, block, and glandular secretion were significantly associated in normal subjects. This suggested that HTS induced pain, nasal discharge, and these two sensations. This is consistent with our previous findings (12, 19, 23, 26, 35) and the hypothesis that HTS stimulated type C neurons that convey these sensations and have axon responses that stimulate glandular exocytosis. Results such as these indicate that subjective descriptors such as drip, block, "fullness," "congestion," and objective measures such as nasal airflow resistance and nasal airspace volume measured by acoustic rhinometry should not be used synonymously. Instead, the distinct mechanisms leading to conscious, subjective sensations must be distinguished from mucosal pathologic events such as plasma extravasation and watery discharge, glandular exocytosis of mucinous gels, and mucosal thickening caused by engorgement of venous sinusoids (4). For example, it has been shown that HTS stimulated equal magnitudes of sensations and secretions before and after nasal decongestion with oxymetazoline (35). HTS induced the same sensations of nasal blockage even when the nasal vessels were maximally constricted and the nasal cavity maximally patent (measured by acoustic rhinometry). This suggested that the sensation of blockage could be appreciated despite having no change in nasal patency. This could explain the paradoxical finding of subjective nasal obstruction in subjects without AR with patent, nonobstructed nostrils.
HTS had no effect on albumin exudation (12, 35). However, capsaicin doses that cause pain ratings of 7 to 10 out of 10 can induce vascular leak in subjects with untreated AR (13, 14, 36). These pain levels are not typically encountered in daily life or AR (itch predominates) (6). It is possible that other non-neurogenic mechanisms may regulate plasma extravasation. For example, denervated nasal polyps have a 10-fold higher plasma flux than normal nasal mucosa (37).
Previous studies have implicated neural axon responses in HTS nasal provocations (12). HTS dose responses cause incremental increases in nerve stimulation and sensations of pain, block, and drip (12). Substance P was released in a dose-dependent manner in the first 3 minutes of HTS provocations. Substance P containing neurons innervate glands (4). These glands express substance PeCpreferring neurokinin-1 receptors (12). The maximum glandular secretion and sensations of block and drip occurred between 3 and 5 minutes. Volume changes detected by acoustic rhinometry between the administration of each HTS dose and 5 minutes later were roughly equivalent to the net mass of secretions blown out of the nostrils (35). This sequence of neural stimulation, pain, and substance P release followed by glandular secretion and sensations of nasal obstruction and rhinorrhea may define the nociceptive nerve axon response to hypertonic conditions in human airways. This response may be an instantly recruited mucosal defense mechanism to protect against the effects of inhaled irritants such as smoke (4). Evidence for axon responses in AR may have been detected as the HTS dose-dependent secretion of lysozyme.
Subjects with sinusitis had high levels of subjective complaints but the lowest recovered weight of lavage fluid. The lavage fluid may have become trapped behind swollen nasal structures, rapidly absorbed into the epithelial lining fluid or tissue, and become unavailable for collection. The relationships between lysozyme, drip, block, and weight demonstrated the importance of glandular secretion for symptom production in sinusitis. Albumin, mucin, and block were related, suggesting that some aspect of sinusitis mucosal inflammation led to increased plasma extravasation, possible venous sinusoid engorgement, and goblet and glandular mucous hypersecretion.
CFS was anomalous. Sinus pain and fullness responses were elevated and not significantly different from sinusitis (Table 2). We have previously shown that subjects with CFS have lower pain thresholds (increased tenderness to pressure) over their maxillary, frontal, and ethmoid regions compared with normal subjects or subjects with AR, acute sinusitis, and chronic sinusitis (38). Subjects with CFS and sinusitis could be differentiated by low systemic pain thresholds in CFS but not sinusitis. Lower pain thresholds and potentially lower thresholds for other nociceptive stimuli such as HTS would be consistent with systemic nociceptive dysfunction in CFS. This concept was supported by the HTS dose responses for pain intensity, duration, block and drip. The magnitudes for these sensations were as severe in CFS as in sinusitis and AR. The cause of this neural hyperresponsiveness in CFS remains unclear but was unlikely to be related to the inflammatory mechanisms of sinusitis or AR (24, 39, 40). Mucin and albumin secretion were elevated compared with normal (41) but did not respond to HTS stimulation. Lysozyme and mucin dose responses were not correlated, suggesting glandular dysfunction with decreased serous cell but increased mucous cell exocytosis. The mechanism was not apparent.
The urea results were initially perplexing. Urea has been thought to passively diffuse from the interstitial fluid into the nasal epithelial lining fluid (25). Albumin has a similar pathway for secretion (5). However, urea increased in HTS dose-dependent fashion, whereas albumin did not. This suggested that the vascular permeability and transepithelial exudation, which account for the increased albumin exudation in conditions such as AR, did not play a role in urea transport after neurogenic stimulation. Instead, urea was correlated to lysozyme under these conditions. This implied that serous cells were the primary source of urea when these glands were stimulated by the inflammatory milieu, as in sinusitis, or by HTS provocation. The link between stimulation of pain-conveying neurons and glandular secretion may have been the nociceptive nerve axon response mechanism (12). Serous cell transport of urea may be analogous to secretion of uric acid, the major antioxidant in nasal secretions (42). Specific urea transport mechanisms in this mucosa have not been defined.
This pilot project had several limitations that can be rectified in future investigations. Unequal numbers of subjects were present in each group. However, highly significant differences in the magnitudes of changes from normal and between groups were still recognized after Bonferroni corrections for multiple comparisons. The subjects with AR had moderate to severe acute seasonal symptoms on the days of their provocations. Some also had perennial AR, but this did not appear to alter the nasal response. It was not clear whether the duration of symptoms affected their responses. Longitudinal studies beginning before a short allergen season in monosensitized subjects (e.g., birch pollen) may be needed to identify any progression of changes in neural sensitivities, central perceptions of symptom severity, and axon response mechanisms. As noted previously here, other provocation agents may have had different effects in these syndromes. It was not feasible to design a study that could simultaneously compare responses to methacholine, histamine, or other agents in tandem with HTS in the same subjects. Instead, additional parallel group studies would be required.
This preliminary study demonstrated differences in neurogenic responses for sinusitis, AR, and CFS from control subjects. Each inflammatory state had elevated HTS-induced sensations compared with normal but had distinct patterns of secreted proteins. This HTS model offers a reductionist approach to understand the neurogenic and other mechanisms that contribute to gland, vascular, and neural hyperresponsiveness in airway disorders.
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
REFERENCES
Meyer P, Andersson M, Persson CG, Greiff L. Steroid-sensitive indices of airway inflammation in children with seasonal allergic rhinitis. Pediatr Allergy Immunol 2003;14:60eC65.
Dahl R, Mygind N. Mechanisms of airflow limitation in the nose and lungs. Clin Exp Allergy 1998;28:17eC25.
Svensson C, Andersson M, Greiff L, Persson CG. Nasal mucosal end organ hyperresponsiveness. Am J Rhinol 1998;12:1:37eC43.
Tai C-F, Baraniuk JN. Upper airway neurogenic mechanisms. Curr Opin Allergy Clin Immunol 2002;2:11eC19.
Raphael GD, Meredith SD, Baraniuk JN, Druce HM, Banks SM, Kaliner MA. The pathophysiology of rhinitis: II: assessment of the sources of protein in histamine-induced nasal secretions. Am Rev Respir Dis 1989;139:791eC800.
Nicolson TA, Bevan S, Richard CD. Characterisation of the calcium response to histamine in capsaicin-sensitive and capsaicin-insensitive sensory nerve. Neuroscience 2002;110:329eC338.
Raphael GD, Druce HM, Baraniuk JN, Kaliner MA. Pathophysiology of rhinitis: I: assessment of the sources of protein in methacholine-induced nasal secretions. Am Rev Respir Dis 1988;138:413eC420.
Mullol J, Baraniuk JN, Logun C, Merida M, Hausfeld JH, Shelhamer JH, Kaliner MA. M1 and M3 muscarinic antagonists inhibit human nasal glandular secretion in vitro. J Appl Physiol 1992;73:2069eC2073.
Baraniuk JN, Silver PB, Kaliner MA, Barnes PJ. Perennial rhinitis subjects have altered vascular, glandular, and neural responses to bradykinin nasal provocation. Int Arch Allergy Immunol 1994;103:202eC208.
Riccio MM, Proud D. Evidence that enhanced nasal reactivity to bradykinin in patients with symptomatic allergy is mediated by neural reflexes. J Allergy Clin Immunol 1996;97:1252eC1263.
Riccio MM, Reynolds CJ, Hay DW, Proud D. Effects of intranasal administration of endothelin-1 to allergic and nonallergic individuals. Am J Respir Crit Care Med 1995;152:1757eC1764.
Baraniuk JN, Ali M, Yuta A, Fang S-Y, Naranch K. Hypertonic saline nasal provocation stimulates nociceptive nerves, substance P release, and glandular mucus exocytosis in normal humans. Am J Respir Crit Care Med 1999;160:655eC662.
Sanico AM, Philip G, Lai GK, Togias A. Hyperosmolar saline induces reflex nasal secretions, evincing neural hyperresponsiveness in allergic rhinitis. J Appl Physiol 1999;86:1202eC1210.
Sanico AM, Stanisz AM, Gleeson TD, Bora S, Proud D, Bienenstock J, Koliatsos VE, Togias A. Nerve growth factor expression and release in allergic inflammatory disease of the upper airways. Am J Respir Crit Care Med 2000;161:1631eC1635.
Koskela H, Di Sciascio MB, Anderson SD, Andersson M, Chan HK, Gadalla S, Katelaris C. Nasal hyperosmolar challenge with a dry powder of mannitol in patients with allergic rhinitis: evidence for epithelial cell involvement. Clin Exp Allergy 2000;30:1627eC1636.
Togias A, Lykens K, Kagey-Sobotka A, Eggleston PA, Proud D, Lichtenstein LM, Naclerio RM. Studies on the relationships between sensitivity to cold, dry air, hyperosmolar solutions and histamine in the adult nose. Am Rev Respir Dis 1990;141:1428eC1433.
Anderson SD, Spring J, Moore B, Rodwell LT, Spalding N, Gonda I, Chan K, Walsh A, Clark AR. The effect of inhaling a dry powder of sodium chloride on the airways of asthmatic subjects. Eur Respir J 1997;10:2465eC2473.
Togias AG, Proud D, Lichtenstein LM, Adams GK III, Norman PS, Kagey-Sobotka A, Naclerio RM. The osmolality of nasal secretions increases when inflammatory mediators are released in response to inhalation of cold, dry air. Am Rev Respir Dis 1988;137:625eC629.
Baraniuk JN, Clauw DJ, Gaumond E. Rhinitis symptoms in chronic fatigue syndrome. Ann Allergy Asthma Immunol 1998;81:359eC365.
Fukuda K, Straus SE, Hickei I, Sharpe MC, Dobbins JC, Komaroff A. The chronic fatigue syndrome: a comprehensive approach to its definition and study. Ann Intern Med 1994;121:953eC959.
Wasserfallen JB, Gold K, Schulman KA, Milzman D, Baraniuk JN. Development and validation of a rhinoconjunctivitis and asthma symptom score for use as an outcome measure in clinical trials. J Allergy Clin Immunol 1997;100:16eC22.
Wasserfallen JB, Gold K, Schulman KA, Baraniuk JN. Item responsiveness of a rhinitis and asthma symptoms score during a pollen season. J Asthma 1999;36:459eC465.
Baraniuk JN, Yuta A, Tai C-F, Park Y-J. Neurogenic responses in rhinosinusitis . Am J Respir Crit Care Med 2004;169:A820.
Baraniuk JN, Clauw DJ, Yuta A, Ali M, Gaumond E, Upadhyayula N, Fujita K, Shimizu T. Nasal secretion analysis in allergic rhinitis, cystic fibrosis and fibromyalgia / chronic fatigue syndrome subjects. Am J Rhinol 1998;12:435eC440.
Kaulbach HC, White MV, Igarashi Y, Hahn BK, Kaliner MA. Estimation of nasal epithelial lining fluid using urea as a marker. J Allergy Clin Immunol 1993;92:457eC465.
Baraniuk JN, Naranch K, Ali M, Le U. Pathophysiology of rhinosinusitis: baseline symptoms and secretions in acute sinusitis (S), allergic rhinitis (AR), the nonallergic rhinitis of chronic fatigue syndrome (CFS) and normal subjects . Am J Respir Crit Care Med 2004;169:A820.
van Cauwenberge PB, Wang D-Y, Ingels KJAO, Bachert C. Rhinitis: the spectrum of disease. In: Busse WW, Holgate ST, editors. Asthma and rhinitis, 2nd ed. Oxford: Blackwell Science; 2000. p. 6eC16.
Kaliner MA, Osguthorpe JD, Fireman P, Anon J, Georgitis J, Davis ML, Naclerio R, Kennedy D. Sinusitis: bench to bedside: current findings, future directions. Otolaryngol Head Neck Surg 1997;116:S1eCS20.
Benninger MS, Ferguson BJ, Hadley JA, Hamilos DL, Jacobs M, Kennedy DW, Lanza DC, Marple BF, Osguthorpe JD, Stankiewicz JA, et al. Adult chronic rhinosinusitis: definitions, diagnosis, epidemiology, and pathophysiology. Otolaryngol Head Neck Surg 2003;129:S1eCS32.
Dray A, Urban L, Dickenson A. Pharmacology of chronic pain. Trends Pharmacol Sci 1994;15:190eC197.
Swinscow TDV, Campbell MJ. Statistics at square one. London: BMJ Books; 2002. p. 44eC51.
Gardner MJ, Altman DG, editors. Statistics with confidence: confidence intervals and statistical guidelines. London: BMJ Books; 1989. pp. 1eC140.
Adam G. Visceral perception: understanding internal organs. New York: Plenum Press; 1998.
Coghill RC, McHaffie JC, Yen Y-F. Neural correlates of interindividual differences in the subjective experience of pain. Proc Natl Acad Sci USA 2003;100:8538eC8542.
Baraniuk JN, Ali M, Naranch K. Hypertonic saline nasal provocation and acoustic rhinometry. Clin Exp Allergy 2002;32:543eC550.
Sanico AM, Atsuta S, Proud D, Togias A. Plasma extravasation through neuronal stimulation in human nasal mucosa in the setting of allergic rhinitis. J Appl Physiol 1998;84:537eC543.
Biewenga J, Stoop AE, van der Heijden HA, van der Baan S, van Kamp GJ. Albumin and immunoglobulin levels in nasal secretions of patients with nasal polyps treated with endoscopic sinus surgery and topical corticosteroids. J Allergy Clin Immunol 1995;96:334eC340.
Naranch K, Park Y-J, Repka-Ramirez SM, Velarde A, Clauw D, Baraniuk JN. A tender sinus does not always mean sinusitis. Otolaryngol Head Neck Surg 2002;127:387eC397.
Repka-Ramirez MS, Naranch K, Park Y-J, Clauw D, Baraniuk JN. Cytokines in nasal lavage fluids from acute sinusitis, allergic rhinitis, and chronic fatigue syndrome subjects. Allergy Asthma Proc 2002;23:185eC190.
Repka-Ramirez MS, Naranch K, Park Y-J, Velarde A, Clauw D, Baraniuk JN. IgE levels are the same in chronic fatigue syndrome (CFS) and control subjects when stratified by allergy skin test results and rhinitis types. Ann Allergy Asthma Immunol 2001;87:218eC221.
Naranch K, Repka-Ramirez SM, Park Y-J, Velarde A, Finnegan R, Murray J, Pheiffer A, Hwang E, Clauw D, Baraniuk JN. Differences in baseline nasal secretions between chronic fatigue syndrome (CFS) and control subjects. J Chronic Fatigue Syndr 2002;10:3eC15.
Peden DB, Hohman R, Brown ME, Mason RT, Berkebile C, Fales HM, Kaliner MA. Uric acid is a major antioxidant in human nasal airway secretions. Proc Natl Acad Sci USA 1990;87:7638eC7642.(James N. Baraniuk, Kristi)