Preseptal cellulitis

INTRODUCTION — Preseptal cellulitis (sometimes called periorbital cellulitis) is an infection of the anterior portion of the eyelid, not involving the orbit or other ocular structures. In contrast, orbital cellulitis is an infection involving the contents of the orbit (fat and ocular muscles) but not the globe. Although preseptal and orbital cellulitis may be confused with one another because both can cause ocular pain and eyelid swelling and erythema, they have very different clinical implications. Preseptal cellulitis is generally a mild condition that rarely leads to serious complications, whereas orbital cellulitis may cause loss of vision and even loss of life. Orbital cellulitis can usually be distinguished from preseptal cellulitis by its clinical features (ophthalmoplegia, pain with eye movements, impaired visual acuity, and proptosis) (table 1) and by imaging studies. In cases in which the distinction is not clear, clinicians should treat patients as though they have orbital cellulitis. Both conditions are more common in children than in adults, and preseptal cellulitis is much more common than orbital cellulitis.

The pathogenesis, microbiology, clinical manifestations, diagnosis, and treatment of preseptal cellulitis will be reviewed here. Orbital cellulitis, subperiosteal abscess, and orbital abscess, are discussed separately. Orbital infections caused by fungi, mainly the Mucorales (which cause mucormycosis) and Aspergillus spp and, much more rarely, Mycobacterium tuberculosis, are also presented elsewhere. (See "Orbital cellulitis" and "Mucormycosis (zygomycosis)" and "Epidemiology and clinical manifestations of invasive aspergillosis" and "Tuberculosis and the eye".)

TERMINOLOGY — Preseptal cellulitis and orbital cellulitis involve different anatomic sites, with preseptal cellulitis referring to infections of the soft tissues anterior to the orbital septum and orbital cellulitis referring to infections posterior to it (figure 1). Neither infection involves the globe. (See 'Anatomy' below.)

There is some debate regarding the appropriate terminology for these infections. Some clinicians use the term "periorbital cellulitis" rather than "preseptal cellulitis" or use the terms interchangeably. We prefer the term "preseptal cellulitis" to make a clear anatomic distinction between this infection and the more serious infection, "orbital cellulitis." Orbital cellulitis is sometimes referred to as "postseptal cellulitis"; we favor the term "orbital cellulitis," and will use it throughout this topic.

ANATOMY — Basic familiarity with the anatomy of the eye is fundamental to understanding the pathogenesis, clinical manifestations, and complications of preseptal and orbital cellulitis. The orbit is a cone-shaped structure, lying horizontally, with its apex in the skull. It is surrounded by paranasal sinuses, namely, the frontal (lying superior), ethmoid (medial) and maxillary (inferior) sinuses (figure 2). The orbit is lined by periosteum. The ethmoid sinuses are separated from the orbit by a paper-thin layer called the lamina papyracea, which contains many perforations for nerves and blood vessels as well as some natural fenestrations termed Zuckerkandl's dehiscences. The most common route of infection to the orbit is by extension from the ethmoid sinuses, likely facilitated through these perforations.

The orbital septum is a membranous sheet that extends from the periosteum of the orbit to the tarsal plate and forms the anterior boundary of the orbital compartment (figure 1). As noted above, preseptal cellulitis involves the soft tissues anterior to the orbital septum.

EPIDEMIOLOGY — Preseptal cellulitis is much more common than orbital cellulitis. In two pediatric case series, 94 percent and 87 percent of cases, respectively, were diagnosed as preseptal cellulitis [1,2]; the remainder of cases were diagnosed as orbital cellulitis. Both infections are much more common in children than in adults; accordingly, most of the data regarding these infections come from studies in children.

It is important to distinguish between preseptal and orbital cellulitis because the complications, treatments, and outcomes of the two entities are very different. (See 'Clinical manifestations' below.)

PATHOGENESIS — Whereas the paranasal sinuses are the main source of infection in orbital cellulitis, a number of preseptal cellulitis cases arise from alternative sources. In one retrospective series of 315 children admitted to the hospital with preseptal or orbital cellulitis, sinusitis was the underlying condition in all 18 children with orbital cellulitis but in only 15 percent of those with preseptal cellulitis [1]. However, other studies have found sinusitis to be the most common cause of preseptal cellulitis [3,4].

Other sites from which preseptal cellulitis may arise are the surrounding skin and soft tissues of the face and eyelids. These tissues may become infected following bacteremia or local trauma [5,6] due to insect bites [6], animal bites [7], or foreign bodies. In one retrospective series of 104 patients with preseptal cellulitis admitted to a tertiary care center, roughly 30 percent were thought to originate from each of the following causes: acute dacryocystitis, sinusitis or upper respiratory tract infection, or trauma (including recent eyelid or strabismus surgery) [8]. In another retrospective study that included 262 children with preseptal or orbital cellulitis, trauma (including insect bites) was much more common in patients with preseptal cellulitis than in those with orbital cellulitis (40 versus 11 percent) [2]. Bacteremic seeding of the preseptal space was more common in the era before the availability of the conjugate vaccines for Haemophilus influenzae type b and Streptococcus pneumoniae. It is still occasionally seen in infants or children with bacteremia due to Streptococcus pyogenes. Pseudomonas aeruginosa bacteremia is rarely associated with preseptal cellulitis in immunocompromised patients [9].

MICROBIOLOGY — Data concerning the microbiology of preseptal cellulitis relate to the site of primary infection. Blood cultures are almost always negative except in the unusual cases of bacteremic preseptal cellulitis. Infections arising from the sinuses or the nasopharynx are likely to be caused by S. pneumoniae, Moraxella catarrhalis, and H. influenzae. In cases due to local skin or skin structure infection (impetigo, cellulitis, or hordeolum) or dacryocystitis, the most common organisms are Staphylococcus aureus and Streptococcus pyogenes. The proportion due to methicillin-resistant S. aureus will vary by age and community to some extent and it is useful to have antibiotic resistance profiles from local communities as a guide to initiating empiric antimicrobials. (See 'Choice of outpatient antibiotic' below.)

Although H. influenzae type b was previously a common cause of bacteremic preseptal infection in young children, routine immunization of children with the H. influenzae type b vaccine has caused a sharp decline in the incidence of this species as a cause of preseptal cellulitis [1]. (See "Epidemiology, clinical manifestations, diagnosis, and treatment of Haemophilus influenzae".)

Infrequent causes of preseptal cellulitis include Acinetobacter species [10,11], Nocardia brasiliensis [12], Bacillus anthracis [13], Pseudomonas aeruginosa [9], Neisseria gonorrhoeae [14], Proteus spp [15], Pasteurella multocida [7], Mycobacterium tuberculosis [16], and Trichophyton spp (the cause of "ringworm") [17]. These pathogens can usually be linked to specific exposures.

CLINICAL MANIFESTATIONS — Patients with preseptal cellulitis typically present with unilateral ocular pain, eyelid swelling, and erythema (picture 1).

Chemosis (conjunctival swelling) may occasionally occur in severe cases of preseptal cellulitis but is uncommon. Leukocytosis may also occur in patients with preseptal cellulitis, but is not a sensitive indicator of this infection. The clinical manifestations of preseptal cellulitis should be distinguished from those of orbital cellulitis (table 1 and table 2), which are discussed in greater detail separately. (See "Orbital cellulitis", section on 'Clinical manifestations'.)

Serious complications are rare in preseptal cellulitis. Reported complications include eyelid necrosis and amblyopia associated with delayed resolution of periorbital swelling [8,18]. However, some experts believe that sinusitis-related causes of preseptal and orbital cellulitis represent a continuum and that inappropriately treated preseptal cellulitis can progress to orbital cellulitis. Therefore, clinicians must be vigilant about treatment of preseptal cellulitis. (See 'Diagnosis' below and "Orbital cellulitis", section on 'Diagnosis'.)

DIFFERENTIAL DIAGNOSIS — The main clinical condition to distinguish from preseptal cellulitis is the more serious orbital cellulitis (table 1). This is discussed in detail elsewhere. (See 'Diagnosis' below.)

A local allergic reaction (eg, to a topical ophthalmic antibiotic or to the sting of an insect) can also have a similar appearance to preseptal cellulitis but can usually be differentiated by the exposure history. Blunt trauma to the eye may also cause eye swelling but the history should reveal the diagnosis. Periorbital swelling from environmental allergies, angioedema, or congestive heart failure is usually bilateral (in contrast to preseptal cellulitis, which is usually unilateral).

Other superficial periorbital processes are typically visually distinct from preseptal cellulitis. Hordeolum (stye) and chalazion can be associated with eyelid inflammation, but are usually discrete nodular lesions (see "Eyelid lesions", section on 'Hordeolum (stye)' and "Eyelid lesions", section on 'Chalazion'). Occasionally, an internal hordeolum points to the conjunctival surface of the eye and causes a diffuse rather than a localized swelling. Examination of the underside of the lid for the presence of a hordeolum helps distinguish this entity from more serious causes of eye swelling. Conjunctivitis can rarely cause eyelid edema, but conjunctival findings (eg, injection and irritation) clearly distinguish this from preseptal cellulitis. (See "Conjunctivitis", section on 'Classification and epidemiology'.)

DIAGNOSIS — Preseptal cellulitis is usually a clinical diagnosis, made in a patient with eyelid swelling and erythema once orbital cellulitis has been excluded. Exclusion of orbital cellulitis is dependent on the absence of proptosis, ophthalmoplegia, pain on eye movement, and impairment of visual acuity (table 1). Rarely, imaging may be necessary in uncertain cases.

Clinical evaluation and diagnosis — Preseptal cellulitis is suspected in patients with unilateral eyelid swelling and erythema (picture 1). A history of recent sinusitis (or prolonged/worsening upper respiratory symptoms), insect bite, or local face and/or eyelid trauma is supportive of the diagnosis. The most important aspect of the diagnostic evaluation is distinguishing preseptal cellulitis from the more severe orbital cellulitis (table 1 and table 2), as these different infections have very different clinical implications [2,19-22]. (See "Orbital cellulitis", section on 'Diagnosis'.)

Findings that should raise concern for orbital rather than preseptal cellulitis include ophthalmoplegia with diplopia, pain with eye movement, visual impairment, and proptosis. This is because orbital cellulitis causes swelling and inflammation of the extraocular muscles and fatty tissues within the orbit, but preseptal cellulitis does not. Although chemosis can rarely occur in severe cases of preseptal cellulitis, it is more commonly suggestive of orbital cellulitis. The absence of fever suggests preseptal cellulitis, although patients with orbital cellulitis are sometimes afebrile. As an example, in a retrospective study that included 262 children, fever occurred more commonly in those with orbital cellulitis than in those with preseptal cellulitis (94 versus 47 percent) [2].

These clinical features usually allow for a reasonable degree of confidence in distinguishing between the two entities. However, the clinical evaluation can be challenging in young children, who may not be able to cooperate with a full examination. In uncertain cases, computed tomography (CT) scanning of the orbits and sinuses is used to distinguish preseptal cellulitis from orbital cellulitis. (See 'Imaging studies' below.)

If the distinction between the two remains uncertain following evaluation, patients should be managed as though they have orbital cellulitis. (See "Orbital cellulitis", section on 'Diagnosis' and "Orbital cellulitis", section on 'Treatment'.)

Laboratory testing — Laboratory testing is generally unnecessary for patients with preseptal cellulitis. We suggest that blood cultures not be routinely performed, as the yield in this setting is extremely low. However, blood cultures are reasonable in very young patients, patients with fever, children who are unimmunized, and in cases in which the possibility of bacteremic preseptal cellulitis or orbital cellulitis cannot be reasonably excluded.

Imaging studies — Contrast-enhanced CT scanning of the orbits and sinuses is helpful for distinguishing between preseptal and orbital cellulitis and is indicated if any clinical signs or symptoms point to orbital cellulitis rather than preseptal cellulitis (table 1 and table 2). CT imaging is also indicated in patients with presumed preseptal cellulitis who exhibit marked eyelid swelling, fever, and leukocytosis, or whose infection fails to show improvement after 24 to 48 hours of appropriate antibiotics.

Imaging findings of preseptal cellulitis demonstrate swelling of the eyelid(s) but no proptosis, no fat stranding of the orbital contents, and no edema of the extraocular muscles. Sinusitis is a common cause of preseptal cellulitis and is present in most cases of orbital cellulitis; if sinusitis is identified, the possibility of orbital cellulitis must be given careful consideration (see "Orbital cellulitis", section on 'Diagnosis'). It is important to note that in some cases of orbital cellulitis, the CT scan abnormalities may be subtle. Imaging studies in patients with suspected orbital cellulitis are discussed in detail separately. (See "Orbital cellulitis", section on 'Imaging studies'.)

TREATMENT

Assessing whether inpatient versus outpatient care is appropriate — Hospitalization is recommended for any of the following:

●Children younger than one year of age

●Children who cannot cooperate fully for an examination

●Patients who are severely ill

●Patients with subtle clinical and/or radiographic findings suggesting that the orbit is involved (table 1)

●Patients who fail to respond to initial outpatient therapy in 24 to 48 hours (see 'Response to therapy' below)

Other patients are typically managed in the outpatient setting.

Patients who require hospitalization — Patients with indications for hospitalization should initially be managed according to recommendations for orbital cellulitis until orbital cellulitis can be excluded. (See "Orbital cellulitis", section on 'Treatment'.)

Patients who can be managed in the outpatient setting — Adults and children older than one year of age with mild preseptal cellulitis and no signs of systemic toxicity can generally be treated as outpatients with oral antibiotics provided that close follow-up can be ensured. Indications for hospitalization are discussed above. (See 'Assessing whether inpatient versus outpatient care is appropriate' above.)

Choice of outpatient antibiotic — Treatment is almost always empiric and based upon knowledge of the common infecting organisms (S. aureus and S. pyogenes when skin and skin structures are thought to be primary sites of infection and H. influenzae, S. pneumoniae, and M. catarrhalis when sinusitis is thought to be the site of infection) and their usual susceptibility patterns, as blood cultures are rarely positive and cultures from the site of infection are difficult to obtain.

The approach to selection of an outpatient antimicrobial regimen depends on the presumed pathogenesis of preseptal cellulitis.

Absence of periorbital skin trauma — In the absence of trauma (eg, insect bite, skin abrasion, blunt trauma), organisms colonizing the sinuses and nasopharynx are likely to be the cause of infection.

●Initial treatment – We suggest empiric monotherapy with amoxicillin-clavulanic acid. Although amoxicillin-clavulanic acid is not active against methicillin-resistant S. aureus (MRSA), MRSA is uncommon in this setting, so initial empiric therapy against MRSA is not indicated.

In children, usual dosing is 45 mg/kg per day divided every 12 hours; dosing for severe infections or when penicillin-resistant S. pneumoniae is a concern (using the 600 mg/5 mL suspension) is 90 mg/kg per day divided every 12 hours. In adults, we administer 875 mg every 12 hours of the immediate release tablet or 2 g every 12 hours of the extended release tablet.

In patients with a nonserious penicillin allergy, a second- or third-generation cephalosporin such as cefuroxime, cefpodoxime, or cefdinir is appropriate. For those with life-threatening allergic reactions, initial empiric therapy should be levofloxacin.

•Cefpodoxime (in children <12 years of age: 10 mg/kg per day divided every 12 hours, usual maximum dose 200 mg; in children ≥12 years and adolescents: 400 mg every 12 hours; in adults: 400 mg every 12 hours)

•Cefuroxime (in children: 20 to 30 mg/kg per day orally divided every 12 hours (maximum daily dose 1000); in adults: 500 mg orally twice daily)

•Cefdinir (in children: 14 mg/kg per day, divided every 12 or every 24 hours, maximum daily dose 600 mg; in adults: 300 mg twice daily)

•Levofloxacin (in children: 10 to 20 mg/kg per day orally divided every 12 to 24 hours (maximum daily dose 500 mg); in adults: 500 mg orally once daily)

●If no clinical improvement – If there is minimal or no clinical improvement within 24 to 48 hours, therapy against MRSA should be added. (See 'Response to therapy' below and 'Presence or recent history of periorbital skin trauma' below.)

Presence or recent history of periorbital skin trauma — In the presence of or recent history of periorbital skin trauma (eg, insect bite, skin abrasion, blunt trauma), S. aureus and S. pyogenes are the common causes of infection.

Due to the prevalence of community-acquired MRSA, patients with a likely S. aureus infection warrant initial empiric therapy against MRSA. Options include:

●Linezolid (children <5 years: 10 mg/kg/dose orally every 8 hours; children 5 to 11 years: 10 mg/kg/dose orally every 12 hours; children ≥12 years and adults: 600 mg orally every 12 hours)

or

●Trimethoprim-sulfamethoxazole (TMP-SMX; in children: 8 to 12 mg/kg per day of the TMP component divided every 12 hours; in adults: one to two double-strength tablets [each tablet contains 160 mg TMP and 800 mg SMX] every 12 hours)

plus one of the following agents (which have activity against group A streptococci and H. influenzae):

•Amoxicillin-clavulanic acid (in children: 90 mg/kg per day divided every 12 hours [using the 600 mg/5 mL suspension]; in adults: 875 mg every 12 hours)

or

•Cefpodoxime (in children <12 years of age: 10 mg/kg per day divided every 12 hours, usual maximum dose 200 mg; in children ≥12 years and adolescents: 400 mg every 12 hours; in adults: 400 mg every 12 hours)

or

•Cefdinir (in children: 14 mg/kg per day, divided every 12 or every 24 hours, maximum daily dose 600 mg; in adults: 300 mg twice daily)

or

•Cefuroxime (in children: 20 to 30 mg/kg per day orally divided every 12 hours (maximum daily dose 1000); in adults: 500 mg orally twice daily)

If drainage will accompany the medical therapy and pathogen identification and susceptibility are known, antibiotics may be changed to target the known causative pathogen.

Clindamycin is not recommended for empiric treatment of preseptal cellulitis because a substantial number of both MRSA and methicillin susceptible S. aureus are resistant to clindamycin.

There are no randomized trials comparing the efficacy of antibiotic regimens, thus, the choice among them depends on factors such as potential allergies, toxicities, drug interactions, cost, availability, and clinician familiarity with the medications.

Topical antibiotics have no role in the treatment of this infection.

Response to therapy — Preseptal cellulitis typically responds rapidly and completely to appropriate antibiotics. Outpatients who worsen over 24 hours or fail to show signs of improvement in 24 to 48 hours should generally be hospitalized, treated with broad-spectrum intravenous antibiotics, undergo a computed tomography scan to evaluate for orbital cellulitis and its complications, and be seen in consultation by an ophthalmologist. (See "Orbital cellulitis", section on 'Treatment'.)

Duration — We generally suggest a duration of five to seven days, but if signs of cellulitis persist at the end of this period, treatment should be continued until the eyelid erythema and swelling have resolved or nearly resolved [23,24]. Treatment recommendations are based on small case series, as there are no controlled trials evaluating the duration of antimicrobial therapy in preseptal cellulitis.

Patients with recurrent preseptal cellulitis — Preseptal cellulitis rarely recurs. When it does, it is usually due to an underlying cause that has not been diagnosed (eg, allergies, chronic sinusitis) [25,26], and treatment involves addressing this issue.

A case series of six patients with recurrent preseptal cellulitis, defined as three or more episodes within a one year period with complete resolution between episodes, identified the underlying cause in two patients as environmental allergies, and in one patient each as recurrent sinusitis, herpes simplex infection, contact dermatitis to cosmetics, and Munchausen’s syndrome [25]. An anatomic abnormality of the ethmoid sinuses may also predispose to recurrent preseptal cellulitis [26].

SUMMARY AND RECOMMENDATIONS

●Preseptal versus orbital cellulitis – Preseptal cellulitis refers to infection of the soft tissues anterior to the orbital septum while orbital cellulitis refers to an infection posterior to it (figure 1). Preseptal cellulitis is generally a mild condition that rarely leads to serious complications, whereas orbital cellulitis may cause loss of vision and, very rarely, even loss of life. Orbital cellulitis can usually be distinguished from preseptal cellulitis by its clinical features (ophthalmoplegia, pain with eye movements, and proptosis) and by imaging studies (table 1); in cases in which the distinction is not clear, clinicians should treat patients as though they have orbital cellulitis.

●Etiology – Preseptal cellulitis arises most commonly from sinusitis or a contiguous infection of the soft tissues of the face and eyelids secondary to local trauma, insect or animal bites, or foreign bodies. (See 'Pathogenesis' above.)

●Microbiology – The causes of preseptal cellulitis are dependent on the underlying pathogenesis (infection of the sinuses versus infection of the skin and skin structures surrounding the eye).

•Staphylococcus aureus (including community-acquired methicillin-resistant S. aureus [MRSA]) and Streptococcus pyogenes are the most common pathogens related to infection of the skin and skin structures (table 3).

•Haemophilus influenzae, Streptococcus pneumoniae, and Moraxella catarrhalis are the most common pathogens related to sinusitis (table 3).

●Clinical presentation – Patients with preseptal cellulitis typically present with unilateral ocular pain, eyelid swelling, and erythema (picture 1). During the initial evaluation, it is critical to distinguish preseptal cellulitis from the more serious orbital cellulitis (table 1 and table 2). Although both preseptal cellulitis and orbital cellulitis typically cause eyelid swelling and erythema, the presence of ophthalmoplegia, pain with eye movements, chemosis, and/or proptosis occur with orbital cellulitis, but not with preseptal cellulitis. (See 'Clinical manifestations' above and 'Clinical evaluation and diagnosis' above.)

●Diagnosis – The diagnosis can usually be made based on these clinical features. However, in uncertain cases, computed tomography scanning of the orbits and sinuses can usually distinguish preseptal cellulitis from orbital cellulitis. (See 'Imaging studies' above.)

●Management – Adults and children older than one year of age with mild preseptal cellulitis can be managed on an outpatient basis, provided that the patient has no signs of systemic toxicity and close follow-up is ensured. Hospitalization is recommended for any of the following and such patients should generally be managed according to the recommendations for orbital cellulitis until it is excluded. (See "Orbital cellulitis", section on 'Treatment'.):

•Children younger than one year of age

•Children who cannot cooperate fully for an examination

•Patients who are severely ill

•Patients with subtle clinical and/or radiographic findings suggesting that the orbit is involved

•Patients who fail to respond to initial outpatient therapy in 24 to 48 hours

●Antibiotic selection  

•In patients with absence of skin trauma in the periorbital region, we suggest monotherapy with amoxicillin-clavulanic acid (Grade 2C). Alternatives include cefpodoxime, cefuroxime, cefdinir, and in cases of severe penicillin allergies, levofloxacin. (See 'Absence of periorbital skin trauma' above.)

•In patients with presence or recent history of skin trauma in the periorbital region, we suggest therapy with activity against S. aureus (including MRSA) and streptococci (Grade 2C). We use therapy with linezolid alone or trimethoprim-sulfamethoxazole plus one of the following agents: amoxicillin-clavulanic acid, cefpodoxime, cefuroxime, or cefdinir. (See 'Presence or recent history of periorbital skin trauma' above.)

●Duration of therapy – We generally administer antibiotics for five to seven days, but if signs of cellulitis persist at the end of this period, treatment should be continued until the erythema and swelling have resolved or nearly resolved. (See 'Duration' above and "Orbital cellulitis", section on 'Treatment'.)

ACKNOWLEDGMENT — The editorial staff at UpToDate would like to acknowledge Michael Barza, MD, who contributed to an earlier version of this topic review.