Acute Bacterial Sinusitis in Children

After completing this article, readers should be able to:The viral upper respiratory tract infection (URI) is the most common illness for which children present to the primary care pediatrician. Approximately 5% to 10% of viral URIs are complicated by acute bacterial sinusitis. Sinusitis results in more than $5.8 billion in health care expenditures in the United States annually, of which $1.8 billion is spent on children younger than 13 years. (1) The diagnosis and treatment of acute bacterial sinusitis present unique challenges to the primary care physician in the acute care setting.The ethmoid and maxillary sinuses develop in the third month of gestation and, although small, are present at birth. The frontal sinuses develop from an anterior ethmoidal air cell and are pneumatized by age 5 or 6 years. The sphenoid sinus starts to become aerated at age 5 years and expands in size into the second or third decade of life.The outflow tract of the maxillary sinus is located at the most superior portion of the medial wall of the sinus (Fig 1). This unfortunate positioning makes gravitational drainage difficult. The clearance of secretions of the sinus is thus dependent on the mucociliary elevator of the mucosa. The maxillary sinus empties via the ostium into the middle meatus of the nasal cavity at a location known as the osteomeatal complex. The maxillary sinus ostia are small, tubular structures with a diameter of only 2.5 mm and a length of 6 mm. The anterior ethmoid and frontal sinuses also empty into the osteomeatal complex in the middle meatus. The posterior ethmoid air cells and the sphenoid sinus drain into the superior meatus.A key concept in understanding the pathogenesis of acute bacterial sinusitis is that the mucosa of the nose and nasopharynx is continuous with the mucosa of the paranasal sinuses. This pseudostratified columnar epithelium clears mucus and other material from the sinus by ciliary action. Any process that affects the nasal mucosa also may affect the sinus mucosa. Unlike the nasal mucosa, which is heavily colonized with bacteria, the paranasal sinuses normally are sterile.The pathogenesis of sinusitis involves 3 key factors: obstruction of the sinus ostia, dysfunction of the ciliary apparatus, and thickening of sinus secretions. The narrow diameter of the sinus ostia allows for easy obstruction. The factors that predispose the ostia to obstruction may be divided into those that result in mucosal swelling and those that result in a direct mechanical effect (Table 1).Viral URI is the most common cause of ostial obstruction in children and frequently precedes the development of sinusitis. Obstruction of the ostia results in a transient increase in pressure in the sinus cavity. As oxygen is depleted, the pressure in the sinus becomes negative relative to the atmosphere. This negative pressure allows for the introduction of bacteria from the nose and nasopharynx into the sinus. When the ostia are obstructed, mucous production by the mucosa continues, resulting in the accumulation of fluid in the sinus cavity and the multiplication of bacteria and the initiation of an inflammatory reaction.In addition to ostial obstruction, dysfunction of the mucociliary apparatus also contributes to the development of sinusitis. During a viral URI, progressive loss of ciliated cells may be observed in the respiratory mucosa.Lastly, the quality and character of sinus secretions play an important role in the pathogenesis of acute bacterial sinusitis. Cilia can beat only in a liquid media, and diseases such as cystic fibrosis result in very thick, viscous secretions that diminish ciliary clearance of fluid and debris from the sinus. Infection of the sinus results in thickening of secretions, compounding this process.The result of a viral URI is that all 3 of these factors are present: ostial obstruction, ciliary dysfunction, and thickening of sinus secretions. The viral URI is the most common predisposing factor to the development of bacterial sinusitis in childhood and accounts for approximately 80% of cases. Allergic inflammation underlies the remaining 20% of cases of acute bacterial sinusitis in children.Knowledge of the microbiology of acute sinusitis has been derived from studies of sinus aspiration. The difficulty in obtaining any sample for culture is that the sinuses are a closed space accessible only through a highly contaminated mucous membrane. Maxillary sinus aspiration in children is a time-consuming procedure that should be performed only by a skilled pediatric otolaryngologist. In this procedure, the nasal mucosa is anesthetized and disinfected with a solution of 10% cocaine. A trocar is passed just below the inferior nasal turbinate across the lateral nasal wall. Aspirated material is sent for Gram stain and quantitative aerobic and anaerobic bacterial culture. The recovery of bacteria in a density of 104 Cfu/mL or more is considered significant growth. A Gram stain finding of at least one organism per high-power field correlates with the isolation of bacteria from nasal secretions in a density of 105 Cfu/mL.Several sinus aspiration studies performed in the 1980s in children with 10 to 30 days of sinus symptoms have provided insight into the microbiology of acute bacterial sinusitis. (2,3) In these studies Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis were responsible for most episodes of sinusitis. S pneumoniae was the most frequently isolated organism, accounting for 40% of isolates, with H influenza and M catarrhalis each accounting for approximately 20% of cases. Less commonly isolated organisms include group A streptococcus, group C Streptococcus, Peptostreptococcus spp, Eikenella corrodens, and Moraxella spp. Anaerobic bacteria are not isolated commonly in patients with acute sinusitis.Because of the invasive nature of sinus aspiration, there has been interest in obtaining specimens for culture from a more accessible site that reflects the microbiology of the sinus. Nasopharyngeal swabs have been studied but unfortunately correlate poorly with sinus aspirate cultures. (2) Endoscopically obtained samples of secretions from the middle meatus near the osteomeatal complex also have been studied. In children, these specimens have poor correlation with sinus aspirates, likely because of the narrow caliber of the nasal passages and the potential for contamination with nasal flora. (4)No sinus puncture studies have been performed since 1984 in children who have acute sinusitis, and yet the microbiology of the nasopharyngeal flora of children has undergone important changes in the past decade. Because the pathogenesis of acute otitis media (AOM) and acute bacterial sinusitis is similar, it is possible to use data derived from studies of tympanocentesis performed in children who have AOM as a surrogate for sinusitis. The middle ear cavity is, in fact, a paranasal sinus.Since the introduction of the 7-valent pneumococcal conjugate vaccine (PCV-7), there has been a significant shift in the pathogens responsible for AOM. In a rural Kentucky clinic, Block et al performed tympanocentesis and culture of middle ear fluid on 381 children ages 7 to 24 months who had AOM in the period before and after the introduction of PCV-7. (5) The proportion of cases of AOM caused by S pneumoniae had decreased significantly from 56% to 41%, whereas that caused by H influenzae had increased from 38% to 57%. The rate of isolation of M catarrhalis remained constant at approximately 10%. Studies performed since the introduction of the 13-valent pneumococcal conjugate vaccine have suggested an even greater decrease in the proportion of AOM cases due to S pneumoniae. (6)In addition to the increase in the rate of isolation of H influenzae in children with AOM, there has been a shift in the susceptibility patterns of this organism. The mechanism by which isolates of H influenzae are resistant to penicillins such as amoxicillin is by the production of β-lactamases. Historically, approximately 20% to 30% of isolates of H influenzae causing sinusitis and AOM in children have produced β-lactamase. Recently, middle ear and nasopharyngeal cultures taken from children in Rochester, New York, have demonstrated beta-lactamase rates as high as 50%. (7) This increase in the rate of resistance has important implications for antibiotic selection because empirical choices will need to include a β-lactamase stable drug.There has been recent controversy in the medical literature over the role of Staphylococcus aureus (including methicillin-resistant strains) in acute sinusitis. Some authors have stated that this pathogen should be considered an important cause of acute bacterial sinusitis. (8) However, these studies have serious methodologic flaws. (9) Most were performed by obtaining cultures from the middle meatus via an endoscope. The middle meatus in healthy children is colonized with S aureus as frequently as 65% of the time. Those authors who performed direct sinus aspirate did not verify decontamination of the puncture site. The isolation of S aureus in these studies likely represents contamination from a colonized nasal mucosa. It is notable also that in the original sinus aspirate studies performed in children, S aureus was not isolated in any of the 50 children enrolled in the study. It is doubtful then that this agent is a significant pathogen in acute bacterial sinusitis. However, S aureus plays an important role in some of the complications of sinusitis.A major task for the clinician who is evaluating a child with respiratory symptoms is to distinguish those who have uncomplicated viral URI from those who have acute bacterial sinusitis. This distinction allows for the identification of children who will benefit from an antimicrobial drug. The clinical course of a viral URI is depicted in Figure 2 and usually begins with a sore or scratchy throat. Children are more likely to have fever with a URI than adults, and when fever occurs, it presents early in the illness and resolves within 1 to 2 days. Fever may be accompanied by constitutional symptoms, such as headache and myalgia.Symptoms of nasal obstruction and discharge predominate during a URI and follow a predictable pattern. The nasal discharge starts as clear and watery, becomes thicker and mucoid, and then is colored and opaque (purulent) for several days. This pattern then reverses, with the nasal discharge becoming watery again or simply drying. Cough and hoarseness also may be prominent in viral URIs. Most uncomplicated URIs last 5 to 10 days, and resolution of symptoms occurs without antimicrobial therapy. Although not all children have complete resolution by day 10, the symptoms usually have peaked in intensity on days 3 to 5 of illness and are improving.The presentation of acute bacterial sinusitis conforms to 1 of 3 predictable patterns (Table 2). The first and most common presentation in children is that of persistent respiratory symptoms. These patients have nasal congestion or nasal discharge, with or without cough, for more than 10 but fewer than 30 days without improvement. The key in distinguishing this presentation from an uncomplicated URI is the lack of improvement of their respiratory symptoms after 10 days. Although patients with uncomplicated viral URI may have residual respiratory symptoms at 10 days, almost always these symptoms are improving.The rhinorrhea may be of any color or character: clear and watery, mucoid, or purulent (thick, colored, and opaque). The cough may be wet or dry in character. When cough is a symptom of acute bacterial sinusitis, it must be present during the day but often is reported to be worse at night. Face pain and headache are rare, although painless morning eye swelling occurs on occasion. Parents of preschool children often will report malodorous breath. The child who has this presentation usually appears only mildly ill, and fever, if it is present, is low grade.The second presentation of acute bacterial sinusitis is that of the onset of severe illness. These patients present with temperatures of at least 38.5°C accompanied by a particular type of nasal discharge, a purulent nasal discharge, for a period of 3 to 4 days. Older children and adults may experience focal face pain, tooth pain, fever, and purulent nasal discharge. The presence of fever for longer than 48 hours distinguishes this presentation from that of an uncomplicated viral URI.The third presentation is one of biphasic or worsening symptoms. In the Scandinavian literature, this is referred to as “double sickening.” These patients have initial symptoms of an uncomplicated viral URI that begins to improve. Then after several days there is a substantial worsening of symptoms with exacerbation of cough or nasal discharge or congestion. A new fever may be present, or fever may recur if it was present at the onset.The physical findings in children who have sinusitis may demonstrate mucopurulent material on the nasal mucosa or in the posterior pharynx. The nasal mucosa itself may be erythematous or boggy and pale, and the oropharynx may be injected. Malodorous breath may be present with sinusitis in the absence of a nasal foreign body or dental disease. Examination of the tympanic membranes may reveal evidence of concomitant AOM or otitis media with effusion. Swelling and discoloration of the eyelids are observed sometimes, and occasionally facial tenderness over the maxillary or frontal sinuses is present. Overall, the physical examination is of limited use in making a specific diagnosis because none of these physical findings distinguishes a viral URI from acute bacterial sinusitis.The diagnosis of acute bacterial sinusitis is almost always a clinical one. Strict application of the clinical criteria (Table 2) that define persistent, severe, and worsening symptoms will distinguish those children with sinusitis from those with viral URI and thus identify a cohort of patients most likely to benefit from an antimicrobial agent. The use of these clinical criteria results in a diagnosis of acute bacterial sinusitis in 6% to 7% of children presenting in a primary care setting with upper respiratory tract symptoms. (10)Although various imaging modalities have been used to assess the paranasal sinuses, imaging is not necessary to confirm the diagnosis of uncomplicated acute bacterial sinusitis. Abnormal findings on imaging studies include complete sinus opacification, mucosal thickening of at least 4 mm, or an air-fluid level. The continuity of the nasal mucosa with the sinus mucosa limits the usefulness of imaging in the diagnosis of sinusitis. It has been demonstrated that both adults and children have significant abnormalities on imaging studies that are performed during an uncomplicated URI.A classic study by Gwaltney et al reported on an experience in adults who had uncomplicated colds. (11) Computed tomographic scans performed within 48 to 96 hours of the onset of symptoms demonstrated abnormalities of the sinuses in more than 80% of patients. In children with viral URI, more than half will have abnormal sinuses on plain radiograph. (12) Accordingly, no imaging study is able to distinguish the inflammation of the sinus mucosa caused by viruses from that due to bacteria.Although imaging studies are not recommended routinely in the diagnosis of sinusitis, a negative radiograph effectively eliminates the diagnosis. Computed tomography and magnetic resonance imaging of the sinuses may be useful when complications of sinusitis are suspected. In this situation, fluid collections that may require surgical drainage may be identified.A skilled pediatric otolaryngologist may perform maxillary sinus aspiration in an outpatient setting. Indications for this procedure include sinusitis unresponsive to multiple courses of antibiotics, orbital or intracranial complications, severe facial pain, and suspected sinusitis in an immunocompromised host in whom unusual pathogens such as fungi may be present.Complications of sinusitis may be divided into those involving the orbit, the central nervous system, or the bone (Table 3). The frontal and ethmoid sinuses are the most common sinuses from which complications arise. The delicate and thin walls of the ethmoid sinuses, called lamina papyracea , allow for spread of infection into the orbit. Orbital complications are the most common and include subperiosteal abscess, orbital cellulitis, and orbital abscess. Signs of orbital infection include eyelid swelling, proptosis, and impairment of extraocular muscle movement. This complication may result from spread of infection through the natural dehiscences between the bones that comprise the medial wall of the orbit (which is also the lateral wall of the ethmoid sinus) or from the development of a subperiosteal abscess of the ethmoid bone.The frontal sinuses share venous drainage with intracranial structures, allowing for infection to develop within the brain and surrounding structures. Intracranial infection may present with headache, seizures, focal neurologic signs, or meningeal signs and can include subdural and epidural empyema and brain abscess. Parenteral antibiotics are necessary for treatment, and surgery is indicated when a drainable fluid collection is present. The bony complication of acute frontal sinusitis is Pott puffy tumor, which is a subperiosteal abscess of the frontal bone.The necessity of treating sinusitis with antibiotics has been controversial. There have been several randomized trials of placebo vs antimicrobial therapy in the treatment of sinusitis in children that have produced conflicting results. Lack of efficacy in 2 of these studies may be explained by underdosing of antibiotics and lack of clear definition of the population of patients included in the studies. (13,14) Two other trials have found consistent benefit to antibiotic treatment when strict clinical criteria are used to select patients with sinusitis. (10,15)In the most recent trial, high-dose amoxicillin-clavulanate was compared with placebo. (10) This study included 58 children ages 1 to 10 years who met stringent criteria for sinusitis based on persistent, worsening, or severe symptoms. A standardized symptom score was used to evaluate treatment effect. Of those children who received antibiotic, 64% were cured or improved vs 32% in the placebo In treatment was in of those placebo vs in the antibiotic of the study This study that antibiotic treatment is when children in an setting are as sinusitis stringent clinical antibiotics used to acute bacterial sinusitis in children are in 4 and amoxicillin-clavulanate should be considered as a agent for the treatment of sinusitis because it has the most in sinus Because the proportion of cases caused by H influenzae is likely and the rate of β-lactamase production by this organism is also the addition of to amoxicillin an over amoxicillin of the amoxicillin for S pneumoniae. Although are more common with amoxicillin-clavulanate than most are and is an agent but should be used at a high in those children at for resistant for those younger than 2 child or received of a agent for treating sinusitis in children is such as or are antibiotics that may be used to sinusitis in children, although are S pneumoniae than As the microbiology of respiratory tract in children is and has demonstrated significant since the introduction of the pneumococcal conjugate S pneumoniae, to decrease in the may be more in the treatment of those children in whom amoxicillin-clavulanate or or a of and may be the increased and of this it is an to the use of antimicrobial in these a is also an agent for children in whom amoxicillin-clavulanate therapy or in the 1 It is not by the and for this The rate of such as or is in patients than other However, the usually is in children, and the of has a on the use of in It was that their use may be there is no and with evidence of complications of sinusitis, or an to antibiotics should be for therapy. uncomplicated sinusitis, or should be used as complications are the should be with the results of culture and are to therapy is in children who have sinusitis and are to therapy with an antimicrobial agent. if present at and a decrease in cough and nasal symptoms occurs within 48 symptoms are not improved within this then clinical is the diagnosis is a antimicrobial should be sinus aspiration may be considered for identification of the of antimicrobial therapy has not been studied patients who have a to the initiation of 10 days of therapy usually is those who at a treating the is symptom an 7 days is such as and have not been found to benefit in children with sinusitis and may be with have a benefit to be recommended children experience of sinus symptoms. The most common cause of such symptoms is viral URI, in those children child such as to cystic an or ciliary may predispose patients to symptoms. The of a child with sinusitis should include with an of quantitative and a for cystic and a of nasal mucosa to assess ciliary and sinusitis not to medical surgical may be sinusitis in children is common in adults and occurs often in children who have the predisposing Children with sinusitis have more of an inflammatory although bacteria may cause acute