Article Text

This article has a correction. Please see:

Download PDFPDF

Pathogenesis and treatment of chronic rhinosinusitis
  1. Andrew J Wood,
  2. Richard G Douglas
  1. Department of Otolaryngology-Head & Neck Surgery, Auckland City Hospital, and Department of Surgery, University of Auckland, Auckland, New Zealand
  1. Correspondence to Dr Richard Douglas, The Department of Surgery, The University of Auckland, Auckland 1142, New Zealand; richarddouglas{at}


Chronic rhinosinusitis (CRS) is one of the most common diseases in western societies, causing significant morbidity and resulting in great financial cost. Some patients suffer persistent or recurrent symptoms despite receiving optimal medical and surgical treatment. The recent publication of revised diagnostic criteria and management guidelines will assist both clinical research and practice. Multiple theories have been advanced regarding the underlying pathogenesis including allergy, bacterial or fungal infection, genetic predisposition and structural anomalies, but at present the majority of cases are still considered idiopathic. Recent studies have shown that traditional laboratory culture techniques may fail to detect microorganisms growing within biofilms or within host mucosal cells. Both bacteria and fungi possess a number of mechanisms for both the evasion and modulation of host immune responses, including the formation of biofilms and the production of superantigens. Historically, treatments such as antibiotics that had been directed at putative causative agents have often been disappointing. There are, however, a broad range of medical and surgical therapies with proven efficacy available to the treating physician. Endoscopic surgical management is evolving rapidly, and there have been pronounced improvements in outcome and reduction in the risk of complications. Recent advances in the understanding of the pathogenesis of this condition have led to some promising therapeutic developments, particularly in respect to topical treatments. Despite improvements in therapy, CRS remains a challenging condition to manage.

View Full Text

Statistics from


Chronic rhinosinusitis (CRS) is one of the most common diseases affecting western societies. According to the Center for Disease Control and Prevention 29.2 million adults in the USA were diagnosed with CRS in 2002.1 The financial cost of sinus disease is correspondingly large with an estimated US$5.8 billion (£3.8 billion, €4.2 billion) directly attributable to its management in the USA in 1996.2 In patients being treated in a specialist clinic, the reduction in quality of life experienced by CRS patients is similar to the reduction reported by patients with moderately severe angina, congestive heart failure or low back pain.3

CRS is defined by symptomatic inflammation of the sinonasal mucosa lasting more than 12 weeks. Patients typically present with symptoms of anterior and/or posterior rhinorrhoea usually in combination with nasal obstruction and often with a reduced sense of smell. Patients may experience a sensation of facial or forehead heaviness, but if headache is the primary presenting feature alternative diagnoses such as atypical migraine should be carefully excluded.4 5

Chronic rhinosinusitis is a condition that is distinct from acute rhinosinusitis in more than just chronicity. The majority of cases of acute rhinosinusitis are caused by infection of the paranasal sinuses developing as a sequela to the common cold. The pathogenesis of CRS is less well understood and almost certainly multifactorial.

This article will summarise current thinking about the aetiology of CRS. A short section outlines the clinical assessment of patients presenting with CRS. This is followed by a review of proven and then more speculative treatments that are available and a suggested schedule for an evidence based optimal treatment regimen is proposed. The principles of surgical treatment are outlined as are the major questions that are likely to direct future research.


In a few cases of CRS an underlying condition is relevant to the pathogenesis, with examples of systemic conditions including cystic fibrosis, primary ciliary dyskinesia, hypogammaglobulinaemia, and aspirin exacerbated respiratory disease (Samter's triad). However, the majority of cases of CRS are currently considered to be idiopathic. CRS may be further classified on the basis of whether polyps are visible in the middle meatus on endoscopic examination or not.5 CRS with nasal polyps is histologically as well as clinically distinct from patients without polyps, being characterised histologically by an eosinophilic infiltrate.6

It is a generally accepted concept that the index event in the development of CRS is obstruction of the sinus ostia by mucosal inflammation. Obstruction disrupts mucociliary clearance and leads to changes in microflora within the sinuses. Recent animal model research in which sinuses are surgically obstructed and then instilled with bacteria have provided strong support for this concept.7


It has been postulated that patients with allergic rhinitis are predisposed to CRS with the associated mucosal oedema causing obstruction of the sinus ostia. There are published data suggesting high rates of allergy in CRS patients—for example, Friedman reported atopic features in 94% in a series of patients undergoing sinus surgery in 1975.8 These data may be influenced by selection bias as they tend to be series from allergy clinics.9 In a series of 192 patients with CRS requiring surgical intervention, the rate of atopic sensitisation was less than that reported in a large number of normal controls.10 Data have also been published implying that worsening allergy does not correlate with CRS, with no increase in CRS reported in pollen sensitised individuals during the pollen season.11

The recently published European Paediatric Ophthalmological Society (EPOS) guidelines conclude that there is an ‘unclear balance of benefit versus harm’ with regard to whether allergy and immune function testing should be pursued in CRS patients.5


Invasive fungal sinusitis, in which fungal infection of the paranasal sinuses spreads to adjacent structures, is fortunately rare and occurs almost exclusively in immunocompromised patients. It is treated with aggressive surgical debridement and systemic antifungals.12 There is a further distinct group who have fungal mycetomas in which fungi grow within a single sinus (usually the maxillary or sphenoid), provoking an inflammatory response in the mucosa of the sinus. Mycetomas are usually only minimally symptomatic, and are frequently diagnosed as an incidental finding in patients who have computed tomography (CT) head scans performed. The role of fungi in idiopathic cases of CRS is much less clear, however.

The debate regarding the role of fungi in idiopathic CRS has followed a similar course to that of bacteria. Assessment of their presence demonstrates fungi to be essentially universal in both CRS and normal subjects,13 and attempts to treat idiopathic CRS with antifungal treatment have not proven successful.14 Conventional laboratory culture data have not made it clear whether the bacterial flora of CRS patients differs from normal controls, with coagulase negative staphylococci, Staphylococcus aureus and Streptococcus species being common organisms in both groups.15 16 Prolonged, apparently appropriate courses of broad spectrum antibiotics often yield disappointing clinical results.17 18 Such culture and therapeutic outcome findings have led some authors to be sceptical about the role of microorganisms in the aetiology of this condition.15

In recent years, however, interest in the role of microorganisms has been rekindled as knowledge of their behaviour in vivo is better understood. It is now recognised that the plating and subsequent culture of microorganisms using conventional laboratory techniques can lead to alteration of their behaviour and consequently significant underestimation of microbial virulence and antibiotic resistance in vivo.


It has recently been reported that both bacterial and fungal biofilms are present more frequently on the mucosa of patients with CRS than in normal controls.19–21 Although definitions vary, the key criteria of a biofilm are the presence of a structured community of microorganisms embedded in an extracellular matrix and associated with a surface, which in CRS is the nasal and sinus mucosa.22 23 Biofims have the potential to provide a reservoir of microorganisms existing in a controlled microenvironment protected from host defenses and antibiotics. The importance of biofilms in the pathogenesis of CRS awaits further clarification.

In the same way that biofilms can afford microorganisms protection from host and extrinsic bactericidal mechanisms, S aureus has been shown to have a tendency to reside in the intracellular compartment, which may provide a further explanation for ongoing inflammation in the face of apparently appropriate immune function and antibiotic treatment.24


It has been proposed that many cases of what had previously been considered to be idiopathic CRS are in fact due to fungal hypersensitivity. There is some evidence that the peripheral blood monocytes of patients with CRS mount a more pronounced inflammatory response to the presence of fungal spores than do normal controls.25

S aureus elaborates a number of powerful virulence factors that may directly alter the immune response in the mucosa. In some individuals, these factors may initiate a chronic inflammatory response characterised by increased leucocyte infiltration and long term histological changes. In the last few years there have been several publications which link the production of S aureus exotoxins with nasal polyposis.26 Recently it has been found that S aureus also produces a cluster of 14 toxins known as the staphylococcal superantigen-like (SSL) proteins that function to inhibit myeloid cell recruitment, complement activation, and Fc mediated phagocytosis.27 Their role in CRS has yet to be elucidated.

Aspirin sensitivity

Some patients with adult onset asthma and nasal polyposis develop a hypersensitivity to aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs). This combination of clinical features has become known as Samter's triad after a Chicago immunologist who described a series of a thousand such patients in the 1960s.28 Aspirin's inhibition of cyclo-oxygenase inhibits synthesis of the relatively anti-inflammatory cytokine prostaglandin E2 (PGE2) while simultaneously increasing synthesis of the cysteinyl leukotrienes, which can cause rhinorrhoea and bronchospasm. This imbalance of prostaglandins and leukotrienes occurs in all patients who take inhibitors of cyclo-oxygenase. Patients with Samter's triad have either an exaggerated release of leukotrienes or an increased sensitivity to their effects, or both these features.

The diagnosis of Samter's triad is usually made on the basis of the worsening of asthma, and nasal symptoms or an anaphylactoid reaction usually within an hour of taking a cyclo-oxygenase inhibitor. Aspirin challenges can be performed in patients who have not taken aspirin or other NSAIDs for many years and so are unable to provide a definite history of their tolerance of these medications. Aspirin challenges should be performed in hospital clinics where facilities for the treatment of severe asthma or anaphylactoid reactions are immediately available.29

Patients with Samter's triad tend to have severe nasal polyposis with a pronounced tendency to postoperative recurrence. They require aggressive treatment, and paradoxically their aspirin hypersensitivity opens up a therapeutic option—aspirin desensitisation. It was first reported more than 60 years ago that patients who could be established on a regular dose of aspirin fared better than before their desensitization; since this time there have been a number of observational studies and some controlled trials of this intervention which have been recently reviewed.30 Aspirin desensitisation can be performed in a rush protocol over the course of a day or as an outpatient over the course of several weeks.

Structural anomalies

Some anatomical variations lead to narrowing of the sinus drainage pathways and it has been suggested predispose patients to CRS. Examples of such abnormalities include the presence of a significant septal deviation, concha bullosa (pneumatisation of the middle turbinate), accessory sinus ostium (through the medial wall of the maxillary sinus posterior to the natural ostium), atypical ethmoid cells that narrow natural drainage pathways (eg, Haller cells), and lateralised middle turbinate. However, there is little convincing evidence to suggest that normal sinonasal anatomical variants such as those listed predispose to the development of CRS.31 32 Their correction is usually carried out in order to permit better surgical access.


There are a handful of clearly defined inherited disorders, most notably cystic fibrosis and Kartagener's syndrome, that strongly predispose to the development of CRS. Genetic variation may play a role in the remaining idiopathic cases. There are some epidemiological data suggesting a heritability of nasal polyposis,33 and various HLA haplotypes have been implicated in increasing the risk of developing nasal polyposis by as much as fivefold.34


Patients with CRS usually present with chronic rhinorrhoea associated variably with nasal obstruction, postnasal drip, facial discomfort or headache and hyposmia.

There are few external signs, but nasal endoscopy may reveal the presence of polyps, mucus or pus in the middle meatus. The mucosa covering the middle and inferior turbinates is usually swollen and paler in colour than it is in health.

Most patients usually undergo a trial of medical treatment before they undergo further investigations. Radiological investigations are indicated when medical treatment has not been successful. Plain films no longer have a significant role in the management of this condition. A CT sinus series assists in the diagnosis, reveals the disease extent, and details any anatomical abnormalities that may impact upon surgery (figure 1).

Figure 1

Computed tomography scan is the radiological investigation of choice in chronic rhinosinusitis (CRS) and typical coronal sections are shown. (a) Normal—note the patent nasal airway and the healthy, non-inflamed mucosal lining. (b) CRS without polyps—note the thickened, inflamed mucosa and the obstruction of the osteomeatal complex. (c) CRS with polyps—note the large polyps obstructing the entire nasal airway and the opacified maxillary and ethmoid sinuses.

In certain cases investigations looking for underlying disorders are indicated. Patients who have purulent sinopulmonary disease should have serum immunoglobulin values measured, and cystic fibrosis and primary ciliary dyskinesia disorders need to be excluded when nasal polyps are recognised in childhood. Unilateral polyps or unilateral changes on CT scan require exclusion of neoplasia by biopsy and often further radiology. In this situation magnetic resonance imaging (MRI) scans are sometimes required to distinguish soft tissue from mucus, but MRI scans are generally not helpful in the management of uncomplicated sinusitis.

It is not our practice to request skin prick tests on patients with CRS because evidence for the role of aeroallergen sensitivity in the pathogenesis of this condition is tenuous.5


It is generally accepted practice to treat patients with uncomplicated CRS initially with medical therapy, and to offer surgical therapy only when medical options have been exhausted or when complications have arisen.5

Our standard medical treatment regimen consists of a 3 week course of prednisone and an antibiotic (usually doxycycline or roxithromycin). Topical treatment with saline lavage delivered by a sinus rinse device and topical corticosteroid spray is begun concurrently and continued until review at 2 or 3 months. If the patient improves sufficiently then topical treatments are continued indefinitely. If there is insufficient improvement in symptoms a CT sinus scan is requested and surgery planned according to the extent of the disease on the scan. In a recent retrospective review of 145 patients with CRS who had been treated using a similar protocol and followed up for between 3 and 17 months, 51% were considered to have been successfully treated, while 18% exhibited partial improvement, and 31% subsequently underwent surgical treatment.35


There are a number of clinical studies that demonstrate some efficacy of oral antibiotics in CRS. Unfortunately, there is disagreement as to the most appropriate antibiotic to use or on how long antibiotics should be administered for.17 36 Particular interest has focused on macrolide antibiotics due to their dual therapeutic action, providing immunomodulation as well as antibiosis.37 A randomised placebo controlled trial of 3 months of roxithromycin treatment in patients with CRS without nasal polyps showed a significant improvement in symptoms in those who received this group.36

Nasal lavage

Regular irrigation of the nose is an inexpensive and essentially risk-free treatment, and there is good evidence supporting its efficacy in the management of CRS.38 There is also some evidence to support the addition of antimicrobial agents such as baby shampoo or mupirocin ointment to the lavage solution when mucosal crusting is a problem.39 40


The powerful anti-inflammatory action of corticosteroids affords them a major role in the treatment of CRS and there are a number of studies supporting the use of both intranasal and systemic corticosteroids in all subgroups of CRS.

A double blind study of 167 patients with CRS without polyps that had failed antibiotic treatment was undertaken where patients were randomised to either placebo or twice daily budesonide nasal spray.41 This study showed a pronounced and highly significant reduction in both subjective and objective measures of CRS. There are also data showing efficacy in nasal polyposis. In one study 310 patients with polyp disease were randomised into three groups receiving either twice daily mometasone furoate, once daily mometasone furoate with once daily placebo, or twice daily placebo.42 Again significant improvements were seen in both subjective and objective measures, including nasal polyp grade. The best results obtained were seen in those receiving twice daily treatment.

While the value of intranasal steroids in the management of CRS is clear there is obviously a wide range of drugs and doses available. Is there an advantage of one intranasal steroid over another? When considering this question one needs to weigh not only efficacy and local side effects (most notably epistaxis), but in the context of potentially life long use perhaps of greatest significance is the systemic bioavailability. There are few good trials directly comparing relative efficacy and side effects of the available products, although it would seem that twice daily dosing may be more efficacious than once daily.42 43

Unsurprisingly, systemic corticosteroids are also effective in the treatment of CRS although clearly their far greater toxicity limits their use to courses of short duration. A 2 week course of 50 mg prednisone has been shown to produce significant subjective, clinical and radiological improvements in CRS with polyps.44

Other medical treatments

Antihistamines are a common and effective treatment for allergic rhinitis,45 a condition that appears to commonly co-exist with CRS.10 As such there is some evidence that, in the context of CRS, antihistamines may provide some symptomatic relief when symptoms of allergic rhinitis such as frequent sneezing exist.46 There is, however, no evidence to support the routine use of antihistamines in CRS.

Little has been published in relation to the use of decongestants in the context of CRS. One might expect that they will be of some short term symptomatic help, particularly during acute exacerbations of symptoms, but the clinician must remain mindful of the obvious temptation for more sustained use with the attached danger of rebound nasal congestion or even rhinitis medicamentosa.


There has been a rapid evolution in sinus surgery techniques over the last few decades. The greatest advance has been in the method of intraoperative visualisation, from use of the headlight to the microscope and then endoscopes used with high definition video cameras. CT image guidance, in which a computer assisted optical tracking system superimposes the position of the probe tip on the patient's CT scan, has enabled very technically challenging procedures to be undertaken.47 These technical advances have led to a notable improvement in patient outcomes, so that the rate of recurrence following sinus surgery has fallen significantly.

The technical improvements described along with pronounced variations in surgical technique leads to a wide variation in published outcome data. A study looking specifically at endoscopic appearances and quality of life measures before and after functional endoscopic sinus surgery (FESS) concluded that significant improvements in both measures were seen.48 FESS is a relatively safe procedure with rates of major complications such as orbital injury, major haemorrhage or CSF leak reported at well under 1% in a large contemporary study.49

The goals of surgery are to remove polyps, pus or debris from the sinuses and to establish patency of the sinus ostia. It is our experience that the mucosa frequently returns to a state of health when ostial patency is achieved, and that surgical failure is usually associated with persistence or recurrence of ostial occlusion. Care is taken not to strip the mucosa as denuded bone will be covered by scar tissue which crusts and not by ciliated respiratory epithelium.

The extent of surgery is largely determined by the extent of the mucosal inflammation seen on CT scan. When inflammatory disease is limited to the region of the maxillary sinus, enlargement of the sinus ostium may be all that is required to bring about resolution of the problem. However, many patients have widespread mucosal inflammation and it is our practice to open the sinuses widely in such cases.

It is important to continue medical treatment postoperatively. We currently recommend that saline lavage and topical corticosteroid sprays are used for at least a year after endoscopic sinus surgery and indefinitely in cases of severe polyposis or after revision surgery.

Some patients experience persistence or recurrence of symptoms after FESS. The most common cause for failure is obstruction of the frontal recess by mucosal inflammation. This often settles with repeated courses of medical treatment. However, if the conservative approach is not successful, this problem can be addressed surgically by drilling out the floor of the frontal sinus and the frontal intersinus septum. This extensive procedure is known eponymously as an endoscopic Lothrop after the surgeon who pioneered this technique by an external approach.50


CRS is an inflammatory disorder that produces a major health and social burden to western societies. We advocate the prescribing of a course of medical treatment as shown in box 1 as initial therapy. In those patients who do not respond or respond only transiently we recommend surgical intervention, with the nature of the procedure tailored to the radiological findings.

Box 1 Suggested medical regimen

  • Saline lavage twice daily

  • Intranasal steroid spray twice daily

  • Roxithromycin 300 mg or doxycycline 200 mg orally once daily for 3 months

  • Prednisone 0.5 mg/kg orally one daily tapering over 3 months

The nature of the inflammatory response in CRS and how it varies across the disease spectrum has been reasonably well characterised. Unfortunately, there is still uncertainty as to how the inflammation is initiated and why it tends to persist. While generalised suppression of the inflammatory response with topical or systemic glucocorticoids provides the most effective treatment, longer courses of macrolide antibiotics have been shown to have a therapeutic effect. Microorganisms are believed to provoke the inflammatory response, but currently available medical treatments may fail because bacteria adopt defence mechanisms such as the forming of biofilms and growing within host cells. We believe that improved understanding of factors such as biofilms and bacterial virulence factors may herald more focused and thus more efficacious therapies.

Multiple choice questions (true (T)/false (F); Answers after the references

1. The dominant inflammatory cell in nasal polyposis is:

  1. Neutrophil

  2. Eosinophil

  3. Basophil

  4. Lymphocyte

2. The following conditions potentially predispose to CRS:

  1. Churg Strauss

  2. Systemic lupus erythematosus

  3. Kartagener's syndrome

  4. Cystic fibrosis

3. Symptoms must be present for how many weeks to make a diagnosis of CRS:

  1. 6

  2. 8

  3. 10

  4. 12

4. The following are common organisms in CRS:

  1. Escherichia coli

  2. Klebsiella

  3. Staphylococci

  4. Clostridium

5. The following treatments have proven efficacy in CRS:

  1. Regular saline lavage

  2. Antihistamines

  3. Decongestants

  4. Steroid nasal spray

Key learning points

  • Chronic rhinosinusitis (CRS) is a major cause of morbidity and financial cost in western societies.

  • Diagnostic criteria are clearly defined. Headache is usually not the primary presenting symptom.

  • The pathogenesis remains unclear and treatment is currently not always effective.

  • Medical treatment is used initially in most uncomplicated cases.

  • When medical treatments fail or complications occur, then surgery is considered. Endoscopic techniques have improved significantly in recent years.

Current research questions

  • Are biofilms central to the disease process? Will antibiofilms therapies be effective?

  • Are there bacteria other than Staphylococcus aureus present in CRS mucosa?

  • What is the clinical significance of nasal polyps?

  • What is the most useful way to subclassify CRS?

  • By what mechanism do corticosteroids exert their therapeutic effect in this condition?

Key references

▶ Fokkens W, Lund V, Mullol J. EPOS 2007: European position paper on rhinosinusitis and nasal polyps. Rhinology 2007;45:1–139.

▶ Wallwork B, Coman W, Mackay-Sim A, et al. A double-blind, randomized, placebo-controlled trial of macrolide in the treatment of chronic rhinosinusitis. Laryngoscope 2006; 116:189–93.

▶ Lund VJ, Black JH, Szabo LZ, et al. Efficacy and tolerability of budesonide aqueous nasal spray in chronic rhinosinusitis patients. Rhinology 2004;42:57–62.

▶ Hissaria P, Smith W, Wormald PJ, et al. Short course of systemic corticosteroids in sinonasal polyposis: a double-blind, randomized, placebo-controlled trial with evaluation of outcome measures. J Allergy Clin Immunol 2006;118:128–33.

▶ Smith TL, Mendolia-Loffredo S, et al. Predictive factors and outcomes in endoscopic sinus surgery for chronic rhinosinusitis. Laryngoscope 2005 Dec;115:2199–205.


  1. A (F); B (T); C (F); D (F)

  2. A (T); B (F); C (T); D (T)

  3. A (F); B (F); C (F); D (T)

  4. A (F); B (F); C (T); D (F)

  5. A (T); B (F); C (F); D (T)


View Abstract


  • Funding The Garnett Passe and Rodney Williams Memorial Foundation, PO Box 577, East Melbourne, VIC 80022. Australia Other Funders: The Garnett Passe and Rodney Williams Memorial Foundation, PO Box 577, East Melbourne, VIC 80022. Australia.

  • Competing interests None.

  • Provenance and peer review Not commissioned; not externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • Correction
    The Fellowship of Postgraduate Medicine