Saturday, December 31, 2005

Periorbital Cellulitis with Intracranial Complications

Zainab Kassim@, Siti Noor Ali Shibramulisi@, Amir Hamzah A. Rahman@, Fadzil A@, Abdul Wahab Jantan@, Adnan O Jwda# Kyaw Tun Wai$, Quazi Manjurul Haque%


A six year old boy with staphylococcus aureus meningo-encephalitis complicating periorbital cellulitis is reported. The presence of cavernous sinus thrombosis and the causes of the accompanying bilateral sixth nerve palsies are discussed by the ophthalmology and the neurosurgical team. The main clinical and micriobiological aspects are discussed by the paediatric and the microbiology team.


A 6-year old boy who was previously well presented to the accident and emergency department with dizziness and vomiting following a fall from a bicycle. There was no loss of consciousness. He was admitted to the surgical ward for observation. On admission, he was fully conscious but febrile. His vital signs were stable. .There was periorbital oedema and chemosis of the right eye. Vision and movement of the right eye were normal. The left eye was normal. Two days later, the fever became high grade and he developed vomiting, headache and drowsiness. He was referred to the paediatric and ophthalmology departments. The attending paediatrician found him to be drowsy with neck stiffness and positive Kernig’s sign. The right eye was proptosed with periorbital oedema, conjunctival injection and restricted ocular movements in all directions. The left eye was normal. Both pupils were reactive to light and there was no relative afferent pupillary defect (RAPD). Fundal examination was normal. A full septic work up excluding lumbar puncture was done. He was accordingly transferred to the paediatric intensive care unit (PICU). The first blood culture taken in the surgical ward grew Staphylococcus aureus sensitive to cloxacillin. He was started on intravenous (i.v.) cefotaxime and cloxacillin. A CT scan of the brain and orbit revealed cerebral oedema and soft tissue swelling around the right orbit. There was no obvious cavernous sinus thrombosis. He was given i.v. mannitol for a total of three days. His full blood picture on admission revealed high white cell count (21,700/L with 90% polymorphonuclear leucocytes). The platelet count was normal. The next day, his level of consciousness improved and the fever subsided. However this time his left eye was noted to be swollen and inflamed. The second blood culture also grew S.aureus. A lumbar puncture requested was refused by the parents. On day 5 of his illness, he developed high-grade fever and became drowsy again. His left eye became more swollen whilst the right eye swelling subsided. There was restricted movement of the lateral gaze. Both pupils were still reactive to light with no RAPD. There was no papilloedema. Intravenous cefotaxime was changed to cefipime. MRI of the brain and orbit which was done at this stage showed no cerebral oedema or cavernous sinus thrombosis. He became apyrexial and remained stable over the next five days.

Lumbar puncture was requested repeatedly during his PICU stay and the parents consented on day 10 of illness when he was already on intravenous cefipime for 5 days. The cerebrospinal fluid (CSF) was clear and colourless with white cell count of 7,000/l. The CSF glucose was 4.2 mmol/l (random blood sugar 6.6mmol/ l), protein 3.5g/dl. Gram stain showed no organism and both latex and culture were negative. On day 11 of his illness, a bruit was heard over his left eye. His lateral gaze restrictions were more obvious. There was no restriction in any other gaze directions. He was then referred to the neurosurgical team. Connective tissue screenings done were normal. He was put on a 5-day course of 60mg oral prednisolone. On the 16th day of illness, he was transferred to the neurosurgical ward. He was later referred to the National University Malaysia Hospital in Kuala Lumpur for cerebral angiogram which was done on day 21 of illness. The result was normal. At this point the bruit was no longer heard over the left eye. He was discharged home. During outpatient follow-up 35 days after his initial admission, there was a slight improvement noted in the left lateral gaze.


The clinical history and examination at presentation to the department clearly suggest central nervous system (CNS) infection; in this case it is most likely meningitis or meningoencephalitis. The CNS infection probably originated from the right orbital cellulitis. Right orbital cellulitis most likely occurred following trauma to the right eye that he sustained one day prior to admission. It is well documented that trauma to the eye predisposes to the development of orbital cellulitis1. The other factors which can lead to orbital cellulitis are sinusitis, dental extraction, post operative and preseptal cellulitis 1, 2

The common organisms involved in orbital cellulitis are Haemophilus inlfuenzae, S. aureus and Streptococcus spp.3. This is compatible with the patient in whom blood cultures taken prior to commencement of antibiotics grew S. aureus. The features of meningism persisted for about 5 days after the onset of clinical symptoms. He then deteriorated on day 5 of illness. However his condition improved when cefotaxime was changed to the fourth generation cephalosporin, cefipime. Cefipime is a new, injectable, x-methoxyimino aminothiazolyl cephalosporin with spectrum of activity which includes many of the Gram-positive and Gram-negative bacteria responsible for severe infection 4. The efficacy of cefipime for the treatment of bacterial meningitis had been previously demonstrated in experimental animal models with group B streptococci, S. aureus, E. coli and Pseudomonas aeruginosa as meningeal pathogens5. Xavier et. al. al. found that cefipime is safe and therapeutically equivalent to cefotaxime for management of bacterial meningitis in infants and children6. In this case, where S aureus meningitis is highly likely, cefipime has led to a tremendous clinical improvement as compared to a combination of cefotaxime and cloxacillin.

Devastating complications including meningitis or meningoencephalitis can develop if prompt and adequate treatment for orbital cellulites is not instituted7. Meningitis or meningoencephalitis carries a high mortality and morbidity rate. In one study on acute bacterial meningitis in children aged between 1 month and 12 years in our hospital, the case fatality rate is 19% with a morbidity rate of 42%8. In another studies, ocular manifestations including squint is one of the morbidities 9. However, whether the patient’s squint will improve with time or not, we will have to wait as we have not found published data on this.

Our biggest hurdle to diagnose meningitis with certainty in the paediatric population in this country is failure to get parental consent for lumbar puncture. This is especially so in the rural population .In this case several attempts were made to persuade the parents to consent for a lumbar puncture. This was only given after the child had been on intravenous cefipime for 5 days. Malik AS found that the number of days after admission does not significantly influence the decision making in giving consent for lumbar puncture10. Factors that influenced parental decision are knowledge about the purpose of lumbar puncture and knowledge about the underlying disease10. Perhaps the opportunity to ask more questions during the course of their son’s illness help to allay the parents’ anxiety on lumbar puncture and have given them some insights on the underlying disease.


We saw the case three days after admission. There was an axial proptosis in the right eye with conjunctival injection and generalised restriction of extraocular muscle movements. Both pupils were reactive to light and there was no RAPD. The left eye was normal. The child showed signs and symptoms of meningitis as explained earlier by the pediatrician. We felt that the child had right orbital cellulitis secondary to the trauma he sustained one day prior to his admission to the hospital. A retrograde infection leading to the involvement of the meninges is a well recognized complication in this situation. The CT scan finding was supportive of such a postulation. However despite the introduction of appropriate antimicrobial therapy and supportive measures his fever did not subside and his conscious level deteriorated. Further more his left eye begins to be affected with conjunctival injection, proptosis, and restricted movements. This development was alarming since the possibility of cavernous sinus thrombosis had to be considered. The changing of the antimicrobial therapy was an appropriate action which resulted in some clinical improvement. The development of bilateral six nerve palsies can be seen in cavernous sinus thrombosis and secondary to increase intracranial pressure (false localizing sign). The patient will be on regular follow-up with our eye clinic; on each visit his visual function will be assessed and also the degree of eye deviation. Only after a reasonable period of stabilization of the degree of the deviation that a surgical option may be considered.

Infections of the orbits are uncommon, but they are potentially devastating infections that can quickly result in blindness, meningitis, or death. The emergency physician must make a rapid and accurate diagnosis and then quickly initiate therapy because visual loss is associated directly with the length of time to definitive treatment. The orbit is a pyramid-shaped bony space in the anterior skull that contains the globe, the blood vessels, and the intraorbital muscles and nerves. The space is bordered on its superior, medial, and inferior sides by the facial sinuses (frontal, ethmoid, sphenoid, and maxillary, respectively). The bony septa separating the orbit from the sinuses are thin and fenestrated, particularly in the medial orbital wall, where the lamina papyracea borders the ethmoid sinus. The anterior border of the orbit is marked by the orbital septum, a fibrous band from the external bony orbit to both eyelids, which effectively separates the preseptal space from the orbital space. The posterior wall of the orbit contains the optic canal and the superior and inferior orbital fissures. The superior orbital fissure connects directly to the cavernous sinus and the intracranial space. The posterior wall is the source of the blood and nerve supply to the orbit. The optic nerve (cranial nerve [CN] II) enters the orbit with the ophthalmic artery through the optic canal. CNs III, IV, and VI; the ophthalmic branch of the trigeminal nerve (CN V1); and the superior ophthalmic vein enter the cavernous sinus after exiting the orbit through the superior orbital fissure. The superior ophthalmic vein provides the main venous drainage for the contents of the orbit. The smaller inferior ophthalmic vein exits the orbit through the inferior orbital fissure with the maxillary branch of the trigeminal nerve (CN V2) and connects with the temporal fossa. Infectious material can be inoculated directly into the orbital soft tissue secondary to trauma, surgery, or orbital foreign bodies. More rarely, orbital infections develop from hematogenous seeding secondary to sepsis or bacterial endocarditis. Orbital cellulitis may or may not progress to a subperiosteal abscess, orbital abscess, or cavernous sinus thrombosis. Cavernous sinus thrombosis is an infectious thrombosis of the cavernous sinus. (The cavernous sinus, a circular venous structure surrounding the pituitary gland, drains blood from both orbits.). Infectious thrombosis most commonly is spread from the orbit via the valveless orbital veins into the cavernous sinus. Once again, this diagnosis is confirmed by CT scan or MRI; however, the physical sign of bilateral posterior orbital disease is highly suggestive. Intracranial infection or cavernous sinus thrombosis can result from any stage of orbital infections. Meningitis is inflammation of the meninges resulting in meningeal symptoms (eg. headache, nuchal rigidity, photophobia) and an increased number of white blood cells in the cerebrospinal fluid (CSF), i.e., pleocytosis. Depending on the duration of symptoms, meningitis is classified as acute or chronic. Acute meningitis denotes the evolution of symptoms within hours to several days, while
chronic meningitis has an onset of weeks to months. Cranial nerve palsies may be observed as a result of increased ICP or the presence of exudates encasing the nerve roots.


We received referral for this patient at day 11 of illness. He has bilateral divergent squint due to bilateral 6th cranial nerves palsies. A bruit was detected over the left eye. There was no evidence of orbital swellings or meningitis. He was transferred to our care when his condition remained static on day 16 of illness. The theoretical and practical background would be discussed here. The abducens nerve (6th cranial nerve) begins at the lower border of pons in the pontomedullary sulcus, 1 cm from the midline. The nerve then ascends through the ventral surface of pons, crossed by anterior inferior cerebral artery, pierces the dura of clivus 2 cm below the posterior clinoids. It passes above the inferior petrosal sinus, beneath the petroclinoid ligament. It travels through cavernous sinus freely, inferolateral to the inferior cerebral artery with attached sympathetic fibers. It exits the cranium at superior orbital fissure to innervate lateral rectus muscle. Possible anatomical aetiologies include Abducens nucleus lesion at the brain stem, lesion of medial longitudinal fasciculus (intranuclear opthalmoplegia), lesion of nucleus prepositus hypoglossi and adjacent medial vestibular nucleus, Abducens nerve lesion and muscular and eyeball lesions. Abducens nerve lesions and muscular and eyeball lesions result in abduction deficit. The examples of the lesions are Graves myopathy (fibrotic medial rectus), Myasthenia gravis, orbital pseudotumour / myositis, orbital trauma (medial rectus entrapment), congenital defects (Duane, Mobius syndrome) and convergence spasm (spasm of near reflex). Causes of localizing infranuclear sixth palsy include pontine syndrome (infarct, demyelination, tumour), cerebello pontine angle lesions (meningioma), clivus lesion (chordoma), disorders of middle cranial fossa and lesions at the cavernous sinus or superior orbital fissure (tumour, inflammation, aneurysm, orbital apex fracture). Possible aetiologies in a child with isolated 6th nerve palsy without other neurological signs are recent viral infection and middle ear infections.

The investigations that may be helpful in double pathology are tensilon test, serum glucose and glycosylated haemoglobin, ESR, contrast CT scan of orbital apex, sella, clivus and cavernous sinus and contrast MRI of cavernous sinus and clivus. Looking back, definite history of head injury leads to consideration of missed injury of brain stem, skull base fracture and untreated diffuse brain injury with marked cerebral oedema which leads to increased intracranial pressure and delayed 6th nerve palsy. The obvious clinical history of meningitis and meningism attracts one to think about basal meningitis and subarachnoid haemorrhage. Eye injury and spreading inflammation point towards cavernous sinus thrombosis and bilateral spread. However, left eye bruit definitely denotes vascular problem (dural AVM, aneurysm, indirect carotid-cavernous fistula or other concomitant connective tissue disorder flare up and orbital pseudotumour). It is not a case of double pathology of isolated 6th nerve palsy. With provisional diagnosis of vascular pathology and differential diagnosis of brain stem disorders, we referred the patient for cerebral angiogram. The result was normal. Our management plans for this patient include observation with close follow up, steroid trial (60mg prednisolone daily for 5 days) and surgical correction after 6-12 months after thorough assessment of lateral rectus and medial restus muscles.


Meningitis due to S aureus accounts for 1-9% of all cases of bacterial meningitis and is associated with mortality rate of 14-77%11,12. It usually is associated with neurosurgical interventions (such as cerebrospinal fluid [CSF] shunts), trauma or underlying conditions such as malignancy, decubitus ulcers, cellulitis, infected intravascular grafts, diabetes mellitus, osteomyelitis or perirectal abscess. S aureus meningitis has 2 different pathogenic mechanisms. In the first form, bacteria are introduced during surgery or by trauma or local spreading (especially coagulase-negative staphylococci) from contiguous infections. Bacteria introduced during surgery cause foreign body infection and subsequent postoperative meningitis. Attachment of S aureus to foreign body involves interaction with proteins of the extracellular matrix: fibrinogen, fibronectin, laminin, thrombospondin, vitronectin, elastin, bone sialoprotein and collagen. S aureus ligands for these host proteins have been characterized, cloned and sequenced. In the second form, hematogenous or spontaneous meningitis, S aureus is disseminated systemically. Infection is more often community acquired, and the incidence of positive blood culture results is higher, as is mortality rate. S aureus attachment to endothelial cells during septicaemia is complex and involves interaction with fibronectin, fibrinogen and laminin. After adhesion, phagocytosis by endothelial cells and induction of tissue factor procoagulant activity occur.

Any localized S aureus infection can lead to bacteraemia. In the preantibiotic era, mortality rate was 82%13,14. Recent studies reported mortality rates between 30% and 40% in non-drug using patients with S aureus septicaemia14. Patient with this type of infection have a lower mortality rate compared to those with hematogenous meningitis, which may be explained by early recognition and less systemic involvement Staphylococcal meningitis is associated with a high mortality rate (about 50% in adult), particularly haematogenous S aureus meningitis (mortality rate 18-56%). The prognosis for CSF shunt infections is more favourable, probably because of earlier recognition15. In one study, 38 of 154 (25%) cases of bacterial meningitis during a 7-year period were nonpneumococcal gram-positive coccal infections; the majority of cases were due to S aureus and S epidermidis15.

Laboratory identifications are: full blood count with differential demonstrates polymorphonuclear leukocytosis and CSF analysis is the diagnostic test of choice for suspected meningitis. The CSF lactate dehydrogenase (LDH) appears to be diagnostic and has a prognostic value in bacterial meningitis. Increase in total LDH is observed consistently in bacterial meningitis, mostly due to increase in fractions 4 and 5, which are derived from granulocytes. LDH fraction 1 and 2, derived presumably from brain tissue, are elevated only slightly in bacterial meningitis but rise sharply in patients who develop neurological sequelae. Leucocyte count in the CSF ranges from 250-100,000/L. Counts above 50,000 raise the possibility of a brain abscess having ruptured into a ventricle. Neutrophils predominate early in infection, but mononuclear cells (lymphocytes, plasma cells, histiocytes) steadily increase as the infection continues. Protein count is higher than 45mg/dL. In most cases the protein count ranges from 100-500mg/dL. Glucose content is usually diminished to below 40mg/dL or to less than 40% of blood glucose level. Gram stain of CSF sediment permits identification of the causative organism in most cases. Other laboratory methods for identification of causative organisms include counterimmunoelectrophoresis (CIE), radioimmunoassay (RIA), latex particle agglutination (LPA), enzyme-linked immunosorbant assay (ELISA), and most sensitive of all – gene amplification by polymerase chain reaction (PCR). Blood cultures should always be obtained. They are positive in 40-60% of patients with Haemophilus influenzae, meningococcal or pneumococcal meningitis, but data are scarce for staphylococcal meningitis16. Blood cultures may provide the only definite clue as to the causative agent if CSF cultures are negative and if more sophisticated diagnostic identification procedures are not readily available.


We believe that the patient suffered from a septic bacteriologic process affecting central nervous system causing cerebral oedema and cranial neuropathies albeit somewhat a rare complication the bilateral sixth nerve palsies due to S aureus meningitis or meningoencephalitis can be due to the inflammatory process of the disease itself. The fall causes trauma to the right eye, which then acts as a nidus of infection (S. aureus) which later spread to the central nervous system. Although we were unable to confirm whether or not he had bacterial meningitis because the cerebrospinal fluid sample was obtained after the patient has been on intravenous cefipime for almost one week, the clinical features strongly suggest CNS infection. Meningitis is still a relatively common condition in Malaysia and a high index of suspicion should be the practical approach in any child presenting with high fever associated with drowsiness. This is particularly important in situations where getting CSF sample is a problem. Prompt treatment with appropriate antimicrobial therapy will prevent further complications of this disease. There is a possibility of some pathology in his cavernous sinus during the acute stage of illness that which fortunately was successfully treated. But we were unable to confirm this suspicion with our radiological imaging. The child however responded fairly well to the treatment and we will continue to monitor his progress in the outpatient clinic.



Jackson K, Baker SR, Arbor A. Clinical implications of orbital cellulites. Laryngoscope 1986, 96: 568-573.
Barone SR, Aiuto LT. Periorbital and orbital cellulites in the Haemophilus influenza vaccine era. J. Ped. Opth. Srabismus 1997, 34: 293-296.
Israele V, Nelson J. Periorbital and Orbital cellulites. Ped. Infect. Dis. J 1087, 6: 404-410.
L. Jauregui, D. Matzke, M. Scott, P. Minns and G. Hageage. Cefipime as treatment for osteomyelitis and other severe bacterial infections. Journal of Antimicrobial Chemotherapy 1993, 32. Suppl. B.: 141-149
Kessler. R.E., M. Bies, R. E. Buck, D.R. Chisholm, T.A. Pursiana, Y.H. Tsai, M. Misiek. K.E.Price, and F.Leitner. Comparison of a new cephalosporin, BMY 28142, with other broad-spectrum b-lactam antibiotics. Antimicrob. Agents Chemother 1985, 27: 207-216.
Xavier S-L, Elizabeth C, Ramiro G, Carmen B, Marcela P, Francisco T, George H. McM. Jr. Prospective randomized comparison of cefipime and cefotaxime for treatment of bacterial meningitis in infants and children. Antimicrob. Agents Chemother 1995, 39: 937-940
Laloyaux P, Vanpee, Gillet JB. Orbital cellulites with abscess formation caused by frontal sinusitis. J Emerg. Med 2000, 18: 253-254.
JA Wahab, AA Fadzil, K Zainab, AM Affendi, M Naznin. A 2-year prospective study on childhood acute bacterial meningitis in Tengku Ampuan Afzan Hospital, Kuantan, Pahang Malaysia. Malaysian J Child Health 2000, 12: 48-54.
Helen B, John de L, Susan H, Catherine P et al. Meningitis in infancy in England and Wales: Follow up at age 5 years. BMJ 2001, 323: 533-536
AS Malik. Consent for lumbar puncture-Factors that influence the decision. Malaysian J Child Health 2000, 12: 55-59


Gordon JJ, Harter DH, Phair JP . Meningitis due to Staphylococcus aureus.M J Med 1995, 78 (6 Pt 1) : 965-70.
Jensen AG, Espersen F, Skinhoj P . Staphylococcus aureus meningitis . A review of 104 Nationwide, Consecutive cases. Arch Intern Med 1993, 153: 1902-8
Millar MR, Keyworth N, Lincoln C . ‘Methicillin resistant’ Staphylococcus aureus in a regional neonatal unit. J Hosp Infect 1987 , 10(2) :187-97.
Quintiliani R, Cooper BW. Current concepts in the treatment of staphylococcal meningitis. J Antimicrob Chemother 1988, 21 Suppl C : 107-14
Roberts FJ, Smith JA, Wagner KR . Staphylococcus aureus meningitis : 26 years’ experience at Vancouver General Hospital. Can Med Assoc J 1983 , 128 (12) : 1418-20
Weinstein MP, LaForce FM, Mangi RJ . Non-pneumococcal gram-positive coccal meningitis related to neurosurgery. J Neurosurg 1977, 47(2) : 236-40


Kanski J J. (eds): Clinical Ophthalmology. 4th ed. Butterworth-Heinemann, 1999, pp 564-56
Westfall CT, Shore JW, Baker AS: Orbital infections. Infectious Diseases. 2nd ed. 1373-7.
Lessner A, Stern GA: Preseptal and orbital cellulitis. Infect Dis Clin North Am 1992 Dec; 6(4): 933-52


Tuesday, December 20, 2005

Managing cellulitis in primary care

The skin normally provides an excellent barrier to infection, both due to providing an impenetrable layer for pathogenic bacteria, and by harbouring non-pathogenic bacteria which prevent the harmful ones from taking over [1],[2]. Yet this barrier can break down, and when this happens, there are plenty of bacteria waiting to take advantage of the opportunity. In many countries, around one-third of the population are asymptomatic carriers of Staph aureus, either constantly or intermittently [3]. It has been estimated that disrupting the skin merely by inserting one silk suture can increase the infectiousness of Staph aureus by a factor of 10,000 [4] .

However, possibly due to enthusiastic management of skin diseases with antibiotics, most patients with atopic dermatitis have been found to harbour Staph aureus on their skin, with higher infection rates than in non-atopic controls [3], and for antibiotic resistance to be high in these bacteria [2] .

The commonest pathogens for skin and soft tissue infections are Staph aureus and beta-haemolytic streptococci, although many other bacteria can cause infections once the skin defences are weakened
[2],[3],[5],[6]. The mechanisms of attack are varied; in addition to the direct effects of colonisation of the body's tissues, they also can produce toxins which may be able to breach the skin's defences and enable the bacteria to penetrate the deeper tissues [2] .

Beta-haemolytic streptococci tend to cause cellulitis which spreads rapidly through the subcutaneous tissues thanks to the action of enzymes such as hyaluronidases produced by the bacteria, whereas staphylococcal infections tend to be more demarcated [3] .

More recently, they have also been shown to produce proteins which can act as 'superantigens'.

These bind directly to antigen-presenting cells and so cause polyclonal T cell activation, which has been implicated in the aggravation of atopic dermatitis, psoriasis and contact dermatitis, as well as the more rare but serious toxic shock syndrome and Kawasaki's syndrome [3] .

What is cellulitis?

Cellulitis is a diffuse suppurative inflammation of subcutaneous tissues [1]. In older people the leg is the usual site, whereas younger people tend to develop the condition in the upper body [7]. Patients admitted to hospital have an average stay of 10 days and 10% of patients suffer long term morbidity afterwards from persistent oedema, recurrent cellulitis and leg ulceration [8]. Of these patients, a portal of entry was found in just 55%. There is also a small risk of more serious complications in young children and immunocompromised adults such as gangrene, metastatic abscesses and sepsis [1].

Risk factors for cellulitis

As we have seen, cellulitis is more likely to follow some damage to the skin. One study found the following conditions in patients with cellulitis [9]:

Diabetes (50% of patients)
Previous cellulitis (50%)
Oedema (45%)
Peripheral vascular disease (40%)

Tinea pedis (32%) Another study showed that about 55% of patients with cellulitis of the leg had a previous minor injury or tinea pedis infection

Periorbital cellulitis

Periorbital cellulitis is usually associated with paranasal sinusitis (43% in one study), trauma (25%) or dental infection (6%) . While the majority of infections are due to Staph aureus [10],[11], some cases in children have been associated with Haemophilus influenzae [12]. Both the number of cases specifically due to this bacterium, and the total number of cases have dropped since the Hib immunisation campaign began [13].

Diagnosing cellulitis

Clinical signsThe diagnosis of cellulitis is based on the clinical signs and symptoms


Localised redness, warmth and tenderness

Less common:

Mild fever and chills


Investigations are rarely helpful in cellulitis. White cell counts are mildly raised if at all [7] . Needle aspiration, Gram stains and blood cultures are usually negative [7],[12],[14],[15],[16], as are lumbar punctures for periorbital cellulitis in children [12]. On average, even culture of swabs grows pathogens in only about half of samples [11]. This means that diagnosis and treatment will always be empirical [15].

Necrotising facsciitis

Necrotising fasciitits is a very rare but serious infection which presents like cellulitis but causes fulminant tissue destruction [17]. It usually follows trauma to the skin, but this may be merely a contusion, burn or insect bite [18]. The infection spreads extremely rapidly and can cause multiorgan failure, adult respiratory distress syndrome [[18], with a 40% mortality rate [[19]. Most infections are polymicrobial with Gram positive and negative, aerobic and anaerobic bacteria, although Group A streptococci are the most common pathogens [18].

It can be difficult to distinguish between necrotising fasciitis and cellulitis. The main features of necrotising fasciitis are:

Oedema of the area with peau d'orange appearance

Blistering and necrosis

Crepitus of the tissues

Pain out of proportion to the clinical findings

Signs of systemic infection may be present
[19], [20] Treatment should be started as soon as possible, and requires surgical debridement of the area and high dose intravenous antibiotics [19],[20],[21].

Management of cellulitis

Management of cellulitis is essentially with oral antibiotics, although rest and elevation of the affected part, moist heat and analgesia may be helpful [1],[7].

Recommended antibiotics are [22],[23]:

Penicillin V 500 mg QDS and Flucloxacillin 500 mg QDS for 7-14 days Cost: £6.60-£13.21

If allergic to penicillin:

Erythromycin 500 mg QDS for 7-14 days Cost :£6.16-£12.32

Co-amoxyclav 500/125 mg TDS for 7-14 days Cost: £15.73-£31.46 Other antibiotics may be of use if local sensitivities mean penicillins and erythromycin are less effective.

Fusidic acid 500 mg daily for 5-10 days was equally effective as flucloxacillin 500 mg TDS with cure rates of over 92%
[3] Cost of fusidic acid: £6.47-£12.94

Azithromycin 500 mg on day 1 and 250 mg daily for days 2-5 was as effective as cephalexin 500 mg BD for 10 days
[24] Cost of azithromycin: £13.43

Azithromycin 1.5 g in divided doses over 5 days was as effective as erythromycin 500 mg QDS for 7 days a
[25] Cost of azithromycin: £13.43

Clindamycin, as an inhibitor of protein synthesis, should be used additionally for severe infections as it will inhibit toxin production
[3] Cost of 300 mg QDS for 7-14 days: £25.48-£50.96

Antibiotic prophylaxis While recurrent attacks are common, there is little evidence that the use of prophylactic antibiotics can reduce recurrence. A study in China found that monthly injections of penicillin made no overall impact on the recurrence rate of cellulitis
[26]. Eradicating Staph aureus in asymptomatic nasal carriers with monthly mupirocin ointment has been shown to reduce the number of skin infections as long as the treatment is continued [3], but the long-term use of prophylactic antibiotics carries a risk of encouraging the development of antibiotic resistance [3].

Chronic leg ulcers While chronic leg ulcers are always contaminated with bacteria, antibiotics should only be given if there are signs of cellulitis in the surrounding tissues


1.Hacker SM. Common infections of the skin. Postgrad Med 1994;96(2):43-52. [Medline abstract]

2.Thestrup-Pedersen K. Bacteria and the skin: clinical practice and therapy update. Br J Dermatol 1998;139:1-3. [Medline abstract]

3.Veien NK. The clinician's choice of antibiotics in the treatment of bacterial skin infection. Br J Dermatol 1998;139:30-36. [Medline abstract]

4.Roth RR, James WD. Microbiology of the skin: resident flora, ecology, infection. J Am Acad Dermatol 1989;22(5 Pt 1):879-883. [Medline abstract]

5.Brook I. Antimicrobial therapy of skin and soft tissue infection in children. J Am Podiatr Med Assoc 1993;83(7):398-405. [Medline abstract]

6.Baddour LM. Cellulitis syndromes: an update. Int J Antimicrob Agents 2000;14(2):113-116. [Medline abstract]

7.Ginsberg MB. Cellulitis: analysis of 101 cases and review of the literature. South Med J 1981;74(5):530-533. [Medline abstract]

8.Cox NH, Colver GB, Paterson WD. Management and morbidity of cellulitis of the leg. J R Soc Med 1998;91(12):634-637. [Medline abstract]

9.Koutkia P, Mylonakis E, Boyce J. Cellulitis: evaluation of possible predisposing factors in hospitalized patients. Diagn Microbiol Infect Dis 1999;34(4):325-327. [Medline abstract]

10. Kanra G, Secmeer G, Gonc EN et al. Periorbital cellulitis: a comparison of different treatment regimens. Acta Paediatr Jpn 1996;38(4):339-342. [Medline abstract]

11.Chang CH, Lai YH, Wang HZ et al. Antibiotic treatment of orbital cellulitis: an analysis of pathogenic bacteria and bacterial susceptibility. J Ocul Pharmacol Ther 2000;16(1):75-79. [Medline abstract]

12.Schwartz GR, Wright SW. Changing bacteriology of periorbital cellulitis. Ann Emerg Med 1996;28(6):617-620. [Medline abstract]

13.Ambati BK, Ambati J, Azar N et al. Periorbital and orbital cellulitis before and after the advent of Haemophilus influenzae type B vaccination. Ophthalmology 2000;107(8):1450-1453. [Medline abstract]

14. Perl B, Gottehrer NP, Raveh D et al. Cost-effectiveness of blood cultures for adult patients with cellulitis. Clin Infect Dis 1999;29(6):1483-1488. [Medline abstract]

15. Lutomski DM, Trott AT, Runyon JM et al. Microbiology of adult cellulitis. J Fam Pract 1988;26(1):45-48. [Medline abstract]

16. Aly AA, Roberts NM, Seipol KS et al. Case survey of management of cellulitis in a tertiary teaching hospital. Med J Aust 1996;165(10):553-556. [Medline abstract]

17.Childs SG. Necrotizing fasciitis: challenging management of a septic wound. Orthop Nurs 1999;18(2):11-17. [Medline abstract]

18.Fontes RA Jr, Ogilvie CM, Miclau T. Necrotizing soft-tissue infections. J Am Acad Orthop Surg 2000;8(3):151-158. [Medline abstract]

19. Adant JP, Bluth F, Fissette J. Necrotizing fasciitis: a life-threatening infection. Acta Chir Belg 1998;98(3):102-106. [Medline abstract]

20. Moss RL, Musemeche CA, Kosloske AM. Necrotizing fasciitis in children: prompt recognition and aggressive therapy improve survival. J Pediatr Surg 1996;31(8):1142-1146. [Medline abstract]

21. Leitch HA, Palepu A, Fernandes CM. Necrotizing fasciitis secondary to group A streptococcus. Morbidity and mortality still high. Can Fam Physician 2000;46:1460-1466.[Medline abstract]

22. British National Formulary , British Medical Association and the Royal Pharmaceutical Society of Great Britain, March 2000

23. Public Health Laboratory Service. Antibiotic guidance template for primary care. 2000.
24. Kiani R. Double-blind, double-dummy comparison of azithromycin and cephalexin in the treatment of skin and skin structure infections. Eur J Clin Microbiol Infect Dis 1991;10(10):880-884. [Medline abstract]

25. Daniel R. Azithromycin, erythromycin and cloxacillin in the treatment of infections of skin and associated soft tissues. European Azithromycin Study Group. J Int Med Res 1991;19(6):433-445. [Medline abstract]

26.Wang JH, Liu YC, Cheng DL et al. Role of benzathine penicillin G in prophylaxis for recurrent streptococcal cellulitis of the lower legs. Clin Infect Dis 1997;25(3):685-689. [Medline abstract]

27. McClure CL. Common infections in the elderly. Am Fam Physician 1992;45(6):2691-2698. [Medline abstract]

JB Medical

Monday, December 19, 2005

Neisseria meningitidis periorbital cellulitis associated with meningitis

First Reported Case of Neisseria meningitidis Periorbital Cellulitis Associated With Meningitis

David V. Chand, MD, Claudia K. Hoyen, MD, Ethan G. Leonard, MD and Grace A. McComsey, MD

From the Department of Pediatrics, Division of Pediatric Infectious Diseases and Rheumatology, Rainbow Babies & Children’s Hospital, Cleveland, Ohio


Cellulitis is a rare manifestation of meningococcal disease. We describe the case of a previously healthy 4-month-old female infant who developed periorbital cellulitis associated with meningococcal meningitis.

Key Words: meningitis • meningococcal disease

Abbreviations: CSF, cerebrospinal fluid

Periorbital cellulitis is often caused by Staphylococcus aureus or Streptococcus pyogenes after local trauma. Before universal immunization with conjugate vaccine, Haemophilus influenzae type b was responsible for 80% of cases of bacteremic periorbital cellulitis.
1 Here we describe a case of periorbital cellulitis associated with meningitis caused by Neisseria meningitidis.

Case Report

A previously healthy 4-month-old Amish female was transferred to our hospital with 2 days of fever, fussiness, decreased oral intake, and decreased wet diapers. Her reported rectal temperature was 40.6°C. She had no cough, congestion, or rhinorrhea. On the day of admission, she vomited 3 times but had no change in stool pattern. She had no sick contacts. She had received her first hepatitis B, polio, H influenzae type b, and diphtheria and tetanus toxoids and acellular pertussis vaccines at 2 months of age. There was no family history of invasive bacterial infection.

At an outside hospital, the patient had a temperature of 38.4°C, pulse of 170 beats per minute, and a respiratory rate of 42 per minute. She was ill-appearing without localizing signs on examination. A serum basic chemistry was normal except for a glucose level of 155 mg/dL. The complete blood count showed a white blood cell count of 8400/mm3, of which 49% were neutrophils, 7% were bands, 33% were lymphocytes, 9% were monocytes, and 2% were atypical lymphocytes; a hemoglobin level of 10.6 g/dL; and a platelet count of 224000/mm3.

Cerebrospinal fluid (CSF) analysis showed a protein concentration of 314 mg/dL; a glucose concentration below the assay range (<1>

At our hospital she was started on meropenem (120 mg/kg per day), the broad-spectrum antibacterial agent used in our intensive care unit as part of a clinical study protocol, and continued on vancomycin. On arrival, the patient was noted to have developed left periorbital edema and erythema. A computed tomography scan of her head showed normal sinuses and left periorbital edema (Fig
1). An MRI demonstrated left periorbital edema, as well as leptomeningeal enhancement. An ophthalmology consultation was obtained, and the examination was also consistent with periorbital cellulitis. The CSF grew N meningitidis serogroup B. Blood and urine cultures remained negative. Total complement function (CH50) was normal. Vancomycin and meropenem were discontinued, and the patient was treated with ceftriaxone (100 mg/kg per day). The patient’s left eye and clinical status improved rapidly, but she remained febrile. A repeat MRI on hospital day 9 revealed a small amount of extra-axial fluid along the floor of the middle cranial fossa bilaterally, suggestive of small sterile effusions.

Her clinical course was also complicated by bilateral hearing loss. Pediatric neurosurgery and otolaryngology consultations were obtained to assist in the management of these complications; no surgical intervention was deemed necessary. She subsequently became afebrile, and an MRI performed 1 week later revealed a smaller fluid collection in the left middle cranial fossa. She was discharged from the hospital after 17 days of intravenous antibacterial therapy. The family and close contacts received the appropriate prophylaxis.


Periorbital cellulitis associated with N meningitidis has rarely been reported. Typically, periorbital cellulitis is caused by skin flora, predominantly S aureus and S pyogenes, after local trauma. It can also result from a localized infection such as conjunctivitis. Less often, this disease can also be associated with bacteremia, historically with H influenzae type b and more recently with Streptococcus pneumoniae.
1,2 Donahue and Schwartz2 have described 70 cases of periorbital cellulitis from 1986 to 1996 at their institution. There were 6 positive blood cultures among the 59 cultures obtained. Five cultures grew Streptococcus species: 2 cases of S pneumoniae and 3 cases of S pyogenes. In contrast to earlier studies in which H influenzae type B was responsible for 80% of the cases of bacteremic periorbital cellulitis, only 1 culture, obtained from a child in 1987 who was not immunized against it, grew this organism. In a similar fashion, with the introduction of the heptavalent pneumococcal conjugate vaccine, there will likely be a decrease in the incidence of periorbital cellulitis caused by S pneumoniae.

There have been 11 reported cases of cellulitis associated with meningococcus.
38 Of these 11 cases, 5 were children (age range: 9 months to 9 years). All 5 cases involving children presented as periorbital cellulitis, and the other 6 cases, involving adults, presented as cellulitis of the limbs, neck, face, and thorax. In contrast to our patient, none of these 11 patients had meningitis, although only 1 other child had a lumbar puncture. Meningococcus was isolated from blood, conjunctival exudates, or the area of cellulitis. However, similar to the other 5 children in the reported cases, our patient had no underlying medical conditions.

The prevalence of asymptomatic nasopharyngeal carriage of N. meningitidis has been found to be 1% to 2% in infants.
9 Our patient did not have conjunctivitis or infection of any structures adjacent to the eye. In addition, her blood culture was negative. A possible explanation for her clinical presentation was that she was colonized with N meningitidis, became transiently bacteremic, and subsequently developed meningitis and periorbital cellulitis. Perhaps meningococcus is a more common pathogen causing periorbital cellulitis than was thought previously. Rarely is a pathogen identified in cases of periorbital cellulitis, and meningococcus is sensitive to most of the antibacterial agents commonly used to treat this disease.

Another possibility is that the conjunctiva served as the portal of entry for our patient’s infection, as has been reported previously.
10 Finally, it is possible, although unlikely, that the periorbital cellulitis was not caused by meningococcus but rather by another pathogen that was sensitive to the broad-spectrum antibacterial agents used to treat our patient. For this reason, intravenous antibiotics were continued for 17 days, well beyond the usual 5 to 10 days required to effectively treat meningococcal disease.

Because meningococcus is a rare cause of periorbital cellulitis and remains sensitive to the antibiotics commonly used to manage this disease, it does not seem to be cost-effective to routinely attempt to identify the responsible pathogen. Blood cultures should be obtained routinely; however, the age of the patient, the presence of comorbid conditions, and the clinical appearance of the patient should still dictate the extent of the laboratory evaluation, particularly the need for CSF examination, performed in cases of periorbital cellulitis.


Address correspondence to Grace A. McComsey, MD, Rainbow Babies & Children’s Hospital, Division of Pediatric Infectious Diseases and Rheumatology, 11100 Euclid Ave, Cleveland, OH 44106. E-mail:

No conflict of interest declared.


Wald ER. Periorbital and orbital infections. In: Long SS, Pickering LK, Prober CG, eds. Principles and Practice of Pediatric Infectious Diseases. 2nd ed. Philadelphia, PA: Churchill Livingstone; 2003

Donahue SP, Schwartz, G. Preseptal and orbital cellulitis in childhood: a changing microbiologic spectrum. Ophthalmology. 1998;105 :1902 –1905; discussion 1905–1906
Newton DA, Wilson WG. Primary meningococcal conjunctivitis. Pediatrics. 1977;60 :104 –106
Sullivan TD, LaScolea LJ. Neisseria meningitidis bacteremia in children: quantitation of bacteremia and spontaneous clinical recovery without antibiotic therapy. Pediatrics. 1987;80 :63 –67
Ferson MJ, Shi E. Periorbital cellulitis with meningococcal bacteremia. Pediatr Infect Dis J. 1988;7 :600 –601
Patrick CC, Furuta GT, Edwards M, Estabrook M, Blake MS, Baker CJ. Variation in phenotypic expression of the Opa outer membrane protein and lipooligosaccharide of Neisseria meningitidis serogroup C causing periorbital cellulitis and bacteremia. Clin Infect Dis. 1993;16 :523 –527
Porras MC, Martinez VC, Ruiz IM, et al. Acute cellulitis: an unusual manifestation of meningococcal disease. Scand J Infect Dis. 2001;33 :56 –59
Cartolano G, Barbier C, Arnoult L, et al. Fatal acute cellulitis due to Neisseria meningitidis. J Clin Microbiol. 2003;41 :3996 –3997
[Abstract/Free Full Text]
Gold R. Clinical aspects of meningococcal disease. In: Vedros N, ed. Evolution of Meningococcal Disease. Vol 2. Boca Raton, FL: CRC Press; 1987

Moraga FA, Domingo P, Banquet N, et al. Invasive meningococcal conjunctivitis. JAMA. 1990;264 :333 –334


Sunday, December 18, 2005

Septicemia Secondary to Vibrio Vulnificus Cellulitis.

Septicemia secondary to Vibrio vulnificus cellulitis

This report published in Communicable Diseases Intelligence Volume 29 Issue, Number 3, reports a case of septicaemia secondary to Vibrio vulnificus cellulitis in an elderly woman which was acquired after wading in a coastal lagoon with a pre-existing superficial leg wound.

Vibrio vulnificus is a naturally occurring, salt-water bacteria found in estuarine and coastal waters worldwide. It prefers low salinity and warm water temperatures for optimum growth. Infection from Vibrio vulnificus is uncommon, although it has been reported from many locations (e.g. southern United States of America, Israel, Republic of Korea, Japan, Taiwan, Spain, Turkey). It can be serious and life threatening, causing septicaemia and wound infections. This paper reports a case of septicaemia secondary to Vibrio vulnificus cellulitis in an elderly woman. The infection was acquired after wading in a coastal lagoon with a pre-existing superficial leg wound. Commun Dis Intell 2005;29:305–307.

Top of page


Vibrio vulnificus is a gram-negative bacillus and part of normal marine flora in estuarine and coastal waters worldwide.1 It has been isolated in waters of low to moderate salinity i.e. 5–25 parts per thousands (ppt)2,3 and in water temperatures of 9–31° C. Vibrio vulnificus is also found in sediment, and filter feeding shellfish such as oysters, mussels, clams, and scallops, and fish that inhabit coastal oyster reefs.4

V. vulnificus illness has been reported worldwide and usually occurs in the warmer months. Gastroenteritis associated with ingestion of uncooked seafood (particularly oysters) contaminated with V. vulnificus is rarely reported. However, primary septicaemia may occur in those with chronic liver disease, haemochromatosis, or immune disorders. The case fatality rate is 50 per cent, increasing to 90 per cent in those with hypotension.1 This clinical syndrome includes fever, chills, hypotension, shock, and metastatic necrotizing cutaneous lesions. Thrombocytopaenia and disseminated intravascular coagulation are common complications. In otherwise healthy people, exposure of superficial wounds to water where the organism is present can result in local wound infection that may progress to cellulitis, necrotizing fasciitis and secondary septicaemia.1,5 The case fatality rate ranges from 20–30 per cent for V. vulnificus wound infections.4

Top of page

Case report

An 83-year-old female had been wading in a coastal lagoon with a pre-existing abrasion on her left lower leg. Two days later she presented to her general practitioner with fever (axillary temperature 39.6 ° C), low abdominal pain, and extreme pain in her left lower leg. The area of abrasion had a motley dark appearance. She was subsequently referred to hospital. Prior to this illness the patient was well, active and independent with no major health issues other than asthma, for which she used a Budesonide inhaler. She had no known history of liver disease or immunosuppression. She did not eat any fresh oysters or seafood leading up to her illness.

On admission to the emergency department the patient was febrile, with a history of rigors, nausea, vomiting, malaise and abdominal pain. She was alert, orientated, and normotensive. Initial treatment for cellulitis included intravenous fluids, penicillin, flucloxicillin, and analgesia. Biochemistry and haematology results were normal (white cell count 10.7 x 109/L; normal range 4.0 – 11.0 x 109/L) except for neutrophil count 9.6 x 109/L (2.0 – 8.0 x 109/L); lymphocytes 0.1 x 109/L (1.0 – 4.0 x 109/L); monocytes 0.9 x 109/L (0.2 – 0.8 x 109/L) and C-reactive protein 8.5 mg/L (<>
Environmental investigation

The lagoon where the patient went wading (lagoon 1) is one of three distinct coastal lagoons, and is separated from the ocean by a sandbar. Water samples were taken from each of the lagoons and adjacent beaches for bacterial analysis and salinity testing (Table). V. vulnificus was isolated from two lagoons, but was not detected in any of the beach samples. All of the beach samples revealed a salinity level of 36.1 parts per thousand (ppt), normal for seawater. Water temperatures recorded for lagoon one fluctuated between 24–28° C at the time of the patient's exposure.


Chemical and microbiological analysis

Lagoon source Salinity
(parts per thousand) Sampling point Faecal coliforms
(per 100 ml) Escherichia coli
(per 100 ml) Vibrio vulnificus
(per 200 ml)
*1 12.6 shallow 24 24 not detected
deep 22 18 detected
2 16.8 shallow 150 150 detected
deep 180 180 detected
3 18.2 shallow 96 96 not detected
deep 8 8 not detected

Source: NSW Health – Division of Analytical Laboratories.

* Patient's wading lagoon.

Other Vibrio infections

The patient's general practitioner also diagnosed a number of other otitis externa infections around the same time. It is possible that these infections were as a result of swimming in the same lagoon. In one case, Vibrio species was cultured from a swab taken (species not identified) when a 14-year-old male presented with an ear infection. Treatment with Augmentin forté and Ciproxin ear drops resulted in a complete recovery.

Top of page

Communicable disease control significance

This case study highlights the need to consider V. vulnificus infection in a differential diagnosis for wound infection, particularly when recreational water activities coincide with growth of the bacteria during the summer months. V. vulnificus infection is also potentially life-threatening for people with pre-existing liver disease and immune disorders. This group may benefit from preventative advice regarding consumption of raw seafood and contact with seawater in the summer months. Rapid progression and severity of disease makes early diagnosis and treatment of V. vulnificus infection crucial for a positive outcome. This infection is not consistently susceptible to aminoglycosides as are other more common aerobic gram-negative bacilli.5 Appropriate antimicrobials include doxycycline, cefotaxime, ceftriaxone, ciprofloxacin or minocycline if V. vulnificus infection is suspected.6

The environmental investigation confirmed the presence of V. vulnificus in local recreational waters with low salinity. It is likely that this bacterium is present during most summers with high water temperatures. It is difficult to quantify the health risk posed by these findings. There is no specific ICD–10 code (International Classification of Diseases – 10th Revision) to allow rapid searching of health databases (in-patient statistics; mortality data). Our local pathology provider upgraded their information system three years ago; there were no other isolates of V. vulnificus in the last three years. Intensive care clinical staff recalled a similar case about 10 years ago. It is equally challenging to communicate a life threatening health risk that is a rare event to a local community that generates income and pleasure from its environment.

Top of page

1. Heymann DL, Ed. Control of Communicable Diseases Manual 18th Edn. Washington, DC: American Public Health Association; 2004.

2. Motes ML, DePaola A, Cook DW, Veazey JE, Hunsucker JC, Garthwright WE, et al. Influence of water temperature and salinity on Vibrio vulnificus in Northern Gulf and Atlantic Coast oysters (Crassostrea virginica). Appl Environ Microbiol 1998;64:1459–1465.

3. Kaspar CW, Tamplin ML, Effects of temperature and salinity on the survival of the Vibrio vulnificus in seawater and shellfish. Appl Environ Microbiol 1993;59:2425–2429 [Review].

4. Strom MS, Paranjpye RN. Epidemiology and pathogenesis of Vibrio vulnificus. Microbes Infect 2000;2:177–188.

5. Carpenter C. Other pathogenic Vibrios. In: Mandell GL, Bennett JE, Dolin R, editors. Mandell, Douglas and Bennett's Principles and Practice of Infectious Diseases. New York: Churchill-Livingstone;1995. p.1945–1948.

Therapeutic Guidelines Limited. 2003. Available Accessed on 29 April 2005.

Top of page

Author affiliations

1. Public Health Director, Central Coast Public Health Unit, Gosford, New South Wales

2. Clinical Nurse Consultant, Infectious Diseases, Central Coast Public Health Unit, Gosford, New South Wales

3. General Practitioner, Coastal Family Practice, Terrigal, New South Wales

4. Environmental Health Officer, Central Coast Public Health Unit Gosford, New South Wales

5. Infectious Disease Officer, Central Coast Public Health Unit, Gosford, New South Wales

Corresponding author: Dr Peter R Lewis, Public Health Director, Central Coast Public Health Unit, PO Box 361, Gosford NSW 2250. Telephone: +61 2 4349 4845. Facsimile: +61 2 4349 4850. Email:

This article was published in Communicable Diseases Intelligence Vol 29 No 3, September 2005.

Australia Dept of Health

Saturday, December 17, 2005

Bullous Cellulitis with Eosinophilia

Bullous "Cellulitis" With Eosinophilia: Case Report and Review of Wells' Syndrome in Childhood

Amy E. Gilliam, MD*, Anna L. Bruckner, MD, Renée M. Howard, MD*, Brian P. Lee, MD, Susan Wu, MD and Ilona J. Frieden, MD*

* Dermatology Pediatrics, University of California, San Francisco, California Department of Dermatology, Stanford University School of Medicine, Stanford, California Children's Hospital and Research Center, Oakland, California


A 1-year-old girl presented with acute onset of edematous erythematous plaques associated with bullae on her extremities and accompanied by peripheral eosinophilia. She was afebrile, and the skin lesions were pruritic but not tender. The patient was treated with intravenously administered antibiotics for presumed cellulitis, without improvement. However, the lesions responded rapidly to systemic steroid therapy. On the basis of lesional morphologic features, peripheral eosinophilia, and cutaneous histopathologic features, a diagnosis of Wells' syndrome was made. Wells' syndrome is extremely rare in childhood, with 27 pediatric cases reported in the literature. Because it is seen so infrequently, there are no specific guidelines for evaluation and management of Wells' syndrome among children. The diagnosis should be considered for children with presumed cellulitis and eosinophilia who fail to respond to antibiotics. Evaluation should include a directed history, physical examination, complete blood count, and stool testing for ova and parasites, to identify potential triggers. Treatment is with systemic steroid therapy unless disease is limited, in which case medium/high-potency topical steroids may be indicated. If systemic features are prominent or disease is chronic (lasting >6 months), then a referral to hematology/oncology should be considered.

Key Words: cellulitis • eosinophilia • eosinophilic cellulitis • Wells' syndrome

Abbreviations: CSS, Churg-Strauss syndrome • HES, hypereosinophilic syndrome • IL, interleukin

Wells' syndrome, or eosinophilic cellulitis, is a rare, recurrent, inflammatory dermatosis of unknown pathogenesis. In 1971, Wells1 described 4 patients with an acute pruritic dermatitis clinically resembling bacterial cellulitis but with histopathologic findings characterized by dermal eosinophilia, phagocytic histiocytes, and the presence of flame figures. He initially called this syndrome recurrent granulomatous dermatitis with eosinophilia but later simplified the name to eosinophilic cellulitis.2

Wells' syndrome is seen more commonly among adults but has been observed among children. Some hypothesize that this syndrome may represent a hypersensitivity response to a circulating antigen.2 Associated precipitants include insect bites, medication reactions, recent immunization, myeloproliferative disorders, malignancies, and infections. We describe a case of a young child with no identifiable triggering factors, and we review the evidence for evaluation and management of these pediatric cases.

Case Report

A previously healthy, 1-year-old girl presented with acute onset of edematous erythematous plaques, with associated bullae, on her lower extremities and left arm (Fig 1). These lesions were pruritic but not painful, and the patient was afebrile. Her parents denied a history of insect bites, ingestion of medications, trauma, or other intercurrent illness. The patient's most recent immunizations had been received 3 months earlier. The patient did not have a history of asthma, and there was no family history of asthma or atopic disease.

The patient was admitted with presumed bacterial cellulitis and was treated with intravenously administered oxacillin, without improvement. Her laboratory studies were significant for an elevated white blood cell count of 30 x 109 cells per L, with peripheral eosinophilia of 48%. After the patient failed to respond to systemically administered antibiotics, examination of vesicle fluid was performed and revealed numerous eosinophils. Subsequently, the diagnosis of probable Wells' syndrome was made. Oral steroid therapy was started at 2 mg/kg, and the patient's cutaneous symptoms improved within 24 hours, leaving residual erythema and hyperpigmentation (Fig 2). Five days after the initiation of oral steroid therapy, a skin biopsy was performed from a persistently indurated area. Histopathologic assessment showed an interstitial infiltrate of histiocytes and waning flame figures, represented by collections of eosinophilic granules surrounded by a palisade of histiocytes (Fig 3).
Oral steroid treatment was tapered over 3 weeks, and a topical triamcinolone preparation was applied to residual lesions twice daily until the lesions resolved. At 1 year, the patient has not experienced recurrent disease.


Wells' syndrome is extremely rare in childhood, with only 27 pediatric cases reported (Table 1). It is characterized by a combination of distinct clinical and histopathologic findings. Classically, patients present with pruritic erythematous plaques, sometimes with associated bullae, that evolve rapidly over 2 to 3 days. These resolve spontaneously over 2 to 8 weeks, leaving residual skin atrophy and hyperpigmentation, resembling morphea.2 There is usually no improvement with antimicrobial therapy; instead, a rapid response to oral corticosteroid treatment is observed, as in our case. It is not uncommon for patients to have recurrent disease, with exacerbations and remissions occurring over several years.
The histopathologic findings are quite specific and are characterized by flame figures, which are composed of eosinophil major basic protein deposited on collagen bundles.3 With resolution, there is a granulomatous phase of histiocytes palisading around the flame figures. Vasculitis is absent, and direct immunofluorescence findings are negative.4,5

Associated laboratory findings include an elevated white blood cell count and peripheral eosinophilia, which is found in up to 50% of cases during the active phase of disease.5,6 The erythrocyte sedimentation rate is elevated for some patients, and there are several reports of associated elevated IgE levels.714 Fever, lymphadenopathy, arthralgias, and other systemic symptoms (such as pulmonary involvement) have been described for Wells' syndrome, and these findings may be indicative of a more severe or progressive course.7

The differential diagnosis of Wells' syndrome includes bacterial cellulitis, Churg-Strauss syndrome (CSS), eosinophilic fasciitis, and hypereosinophilic syndrome (HES) (Table 2). The skin lesions of Wells' syndrome are distinguished from those of bacterial cellulitis by the absence of tenderness and the presence of pruritus, which is often the primary symptom of Wells' syndrome. Lack of warmth, failure to respond to antibiotic therapy, and characteristic histologic findings are the other features that differentiate Wells' syndrome from bacterial cellulitis.

CSS should be considered for patients with persistent peripheral eosinophilia and skin lesions. Although more commonly seen among adults, CSS can present in childhood.1517 This syndrome is characterized by asthma, peripheral eosinophilia, and vasculitis and is associated with autoantibodies to perinuclear antineutrophil cytoplasmic antibody, as well as cutaneous and systemic granulomas. Palpable purpura, tender subcutaneous nodules, and cutaneous infarctions are more often the associated skin findings in CSS, whereas patients present with bullae and vesicular lesions in Wells' syndrome. In both conditions, flame figures can be identified histopathologically, as can peripheral blood and tissue eosinophilia. However, the presence of vasculitis with extensive fibrinoid necrosis of collagen is more suggestive of CSS.2,9
Eosinophilic fasciitis is another condition that can resemble Wells' syndrome. Also seen more frequently among adults but reported in children,
18,19 it presents with acute onset of skin inflammation and resolves with hyperpigmentation and scleroderma-like skin changes. Unlike Wells' syndrome, eosinophilic fasciitis is characterized by arthritis as a prominent symptom, and it follows a more chronic course, with individual lesions requiring months or years to resolve.12 It is distinguishable from Wells' syndrome by the depth of inflammation, with eosinophilic invasion into deeper fascial tissues.

Finally, idiopathic HES also should be considered when the diagnosis of Wells' syndrome is being entertained. This condition is extremely rare in the pediatric age group but has been reported in childhood.20 HES is a lymphoproliferative disorder characterized by overproduction of eosinophils with a predilection to damage specific organs, especially the cardiovascular system. It is defined by sustained eosinophilia (>1.5 x 109 cells per L, lasting for >6 months), with evidence of multiple-organ system involvement, in the absence of parasitic disease, allergic diatheses, or other conditions known to cause eosinophilia. The heart, lungs, central and peripheral nervous systems, kidneys, and gastrointestinal tract can be affected, and the cutaneous findings are similar to those of Wells' syndrome, including erythematous pruritic papules and nodules, urticaria, and angioedema.9,2123 Histopathologically, the skin lesions of HES are nonspecific, and the flame figures and granulomatous infiltrate seen in Wells' syndrome are absent.21

The pathogenesis of Wells' syndrome is not well defined. One hypothesis is that it represents a hypersensitivity mechanism triggered by factors such as infections, drugs, or internal disease. However, in approximately one half of reported cases among children, there is no identifiable precipitating factor.24

Reported precipitants have included bites or stings from ticks, bees, and spiders8,9,12,2528 and infections with mumps, molluscum contagiosum, varicella, and herpes simplex virus.1,24,29,30 There are also several reports of Wells' syndrome associated with bacterial, parasitic, and fungal infections.2,9,14,3133 Numerous medications have been implicated as triggers for Wells' syndrome.13,5,9,3438 Also, several cases of Wells' syndrome occurred after vaccinations,39,40 and it was proposed that the preservative thimerosol was the causative agent in those cases.39 Several cases of Wells' syndrome among adults have been associated with hematologic disorders,9,41 lymphoproliferative malignancies,1,34,42 and carcinoma.34,4345 Zachary et al45 reported a case of Wells' syndrome in a 17-year-old girl with nasopharyngeal carcinoma, which is the only pediatric case of Wells' syndrome associated with malignancy reported in the literature.

One of the key events in disease expression of Wells' syndrome appears to be aberrant and inadequate eosinophil skin homing. Increased interleukin (IL)-5 levels have been observed in Wells' syndrome, and IL-5 not only mobilizes eosinophils from the bone marrow but also promotes homing of eosinophils by altering expression of adhesion molecules. In addition, increased levels of IL-5 appear to induce expression of CD25, the chain of the IL-2 receptor, which enhances eosinophil degranulation and subsequent tissue destruction.4648

Treatment for Wells' syndrome is sometimes unnecessary, because cases often resolve spontaneously. If an infection or other treatable precipitating factor can be identified, then there is often improvement with treatment of the underlying condition.14,30,33,44 However, when no treatable underlying factor can be identified, systemic corticosteroid therapy is used frequently for both adults and children. Most cases resolve after a single course of systemic corticosteroid therapy; when recurrences occur, however, alternative treatments should be considered, to avoid the side effects of chronic systemic steroid therapy.36 Topical steroid treatment has also been reported as successful therapy, both alone2,29,39 and in combination with systemic steroid therapy.5 Specifically, topical steroid therapy alone resolved skin lesions for 2 children, which suggests that topical steroid therapy may be a safe alternative to systemic corticosteroid therapy in the pediatric age group.29,39 Other therapies reported to be successful include various antimicrobial agents,2,5,6,9,10,13,49 colchicine,13 antimalarial drugs, cyclosporine,50 azathioprine,5 interferon-,51 psoralen with ultraviolet A,52 and antihistamines.10,43,53

The small numbers of cases and the fact that most reports are anecdotal make it difficult to draw conclusions regarding whether these therapies are truly effective or these cases resolved spontaneously. However, we think that first-line treatment for children should be systemic corticosteroid therapy, with the addition of topical steroid treatment depending on the extent of disease. A dose of orally administered prednisolone or prednisone of 2 mg/kg per day for 5 to 7 days, with a taper over 2 to 3 weeks, is appropriate. Topical steroid treatment may be used in combination. In cases in which there is limited body surface area involved (15–30%) and an absence of systemic symptoms, it may be prudent to consider medium-potency topical steroid therapy alone, with close follow-up monitoring. This would also be appropriate for recurrent cases of Wells' syndrome identified early, when disease may be limited.

Evaluation should be directed at ruling out other conditions that mimic Wells' syndrome, as well as evaluating possible triggering factors. We recommend a complete history and review of systems, with specific attention to recent medications, vaccinations, insect bites, infections, or illnesses and associated medical problems such as asthma. A thorough physical examination should be performed, with attention to the liver, spleen, and lymph nodes. Stool samples should be sent for ova and parasite testing, and a complete blood count with differential evaluation should be performed. If there is uncertainty regarding the diagnosis, then a skin biopsy should be performed to distinguish between Wells' syndrome and other conditions that mimic it. We do not recommend a complete hematologic evaluation in all cases, because it is rare to see cases of either HES or Wells' syndrome triggered by hematologic or oncologic disorders in childhood.

However, if a child presents with either systemic features, such as fevers, arthralgias, or other organ system involvement, or a chronic course, defined as >6 months of peripheral eosinophilia or recurrences of clinical disease, then a referral to hematology/oncology should be considered.


Accepted Dec 29, 2004.
Address correspondence to Amy E. Gilliam, MD, Department of Dermatology, University of California, 1701 Divisadero St, 3rd Floor, San Francisco, CA 94115. E-mail:

No conflict of interest declared.


Wells GC. Recurrent granulomatous dermatitis with eosinophilia. Trans St Johns Hosp Dermatol Soc.1971; 57 :46 –56[Medline]
Wells GC, Smith NP. Eosinophilic cellulitis. Br J Dermatol.1979; 100 :101 –109
Peters MS, Schroeter AL, Gleich GJ. Immunofluorescence identification of eosinophil granule major basic protein in the flame figures of Wells' syndrome. Br J Dermatol.1983; 109 :141 –148
Bonvalet D, Caron C, Levet R, et al. Wells' syndrome or eosinophilic cellulitis: apropos of 2 cases: review of the literature [in French]. Ann Dermatol Venereol.1983; 110 :899 –907
Moossavi M, Mehregan DR. Wells' syndrome: a clinical and histopathologic review of seven cases. Int J Dermatol.2003; 42 :62 –67
Fisher GB, Greer KE, Cooper PH. Eosinophilic cellulitis (Wells' syndrome). Int J Dermatol.1985; 24 :101 –107
Kamani N, Lipsitz PJ. Eosinophilic cellulitis in a family. Pediatr Dermatol.1987; 4 :220 –224
Anderson CR, Jenkins D, Tron V, Prendiville JS. Wells' syndrome in childhood: case report and review of the literature. J Am Acad Dermatol.1995; 33 :857 –864
Aberer W, Konrad K, Wolff K. Wells' syndrome is a distinctive disease entity and not a histologic diagnosis. J Am Acad Dermatol.1988; 18 :105 –114
Stam-Westerveld EB, Daenen S, Van der Meer JB, et al. Eosinophilic cellulitis (Wells' syndrome): treatment with minocycline. Acta Derm Venereol.1998; 78 :157
Lindskov R, Illum N, Weismann K, et al. Eosinophilic cellulitis: five cases. Acta Derm Venereol.1988; 68 :325 –330
Saulsbury FT, Cooper PH, Bracikowski A, et al. Eosinophilic cellulitis in a child. J Pediatr.1983; 102 :266 –269
Paquet P, Laso-Dosal F, de la Brassinne M. Wells' syndrome: report of 2 cases. Dermatology.1992; 184 :139 –141
Tsuda S, Tanaka K, Miyasoto M, Nakama T, Sasai Y. Eosinophilic cellulitis (Wells' syndrome) associated with ascariasis. Acta Dermatol Venereol.1994; 74 :292 –294
Wang SJ, Yang YH, Lin YT, et al. Childhood Churg-Strauss syndrome: report of a case. J Microbiol Immunol Infect.2000; 33 :263 –266
Lin TL, Wang CR, Liu MF, et al. Multiple colonic ulcers caused by Churg-Strauss syndrome in a 15-year-old girl. Clin Rheumatol.2001; 20 :362 –364
Louthrenoo W, Norasetthada A, Khunamornpong S, et al. Childhood Churg-Strauss syndrome. J Rheumatol.1999; 26 :1387 –1393
Ching DW, Petrie JP. Childhood eosinophilic fasciitis presenting as inflammatory polyarthritis and associated with selective IgA deficiency. Ann Rheum Dis.1991; 50 :647 –648
Huppke P, Wilken B, Brockmann K, et al. Eosinophilic fasciitis leading to painless contractures. Eur J Pediatr.2002; 161 :528 –530
Schulman H, Hertzog L, Zirkin H, Hertzanu Y. Cerebral sinovenous thrombosis in the idiopathic hypereosinophilic syndrome in childhood. Pediatr Radiol.1999; 29 :595 –597
Kazmierowski JA, Chusid MJ, Parrillo JE, et al. Dermatologic manifestations of the hypereosinophilic syndrome. Arch Dermatol.1978; 114 :531 –535
Fujii K, Tanabe H, Kanno Y, et al. Eosinophilic cellulitis as a cutaneous manifestation of idiopathic hypereosinophilic syndrome. J Am Acad Dermatol.2003; 49 :1174 –1177
Tsuji Y, Kawashima T, Yokota K, et al. Wells' syndrome as a manifestation of hypereosinophilic syndrome. Br J Dermatol.2002; 147 :811 –812
Stavropoulos PG, Kostakis PG, Panagiotopoulos AK, et al. Molluscum contagiosum and cryosurgery: triggering factors for Wells' syndrome? Acta Derm Venereol.2003; 83 :380 –381
Nielsen T, Schmidt H, Sogaard H. Eosinophilic cellulitis (Wells' syndrome) in a child. Arch Dermatol.1981; 117 :427 –429
Wood C, Miller AC, Jacobs A, et al. Eosinophilic infiltration with flame figures. Am J Dermatopathol.1986; 8 :186 –193
Clark P, Anderson PC. Eosinophilic cellulitis caused by arthropod bites. Int J Dermatol.1988; 27 :411 –412
Schorr WF, Tauscheck AL, Dickson KB, Melski JW. Eosinophilic cellulitis (Wells' syndrome): histologic and clinical features in arthropod bite reactions. J Am Acad Dermatol.1984; 11 :1043 –1049
Reichel M, Isseroff RR, Vogt PJ, Gendour-Edwards R. Wells' syndrome in children: varicella infection as a precipitating event. Br J Dermatol.1991; 124 :187 –190
Ludwig RJ, Grundmann-Kollmann M, Holtmeier W, et al. Herpes simplex virus type 2-associated eosinophilic cellulitis (Wells' syndrome). J Am Acad Dermatol.2003; 48 (suppl):S60 –S61
Prendiville JS, Jones RR, Bryceson A. Eosinophilic cellulitis as a manifestation of onchocerciasis. J R Soc Med.1985; 78 :21 –22
Van den Hoogenband HM. Eosinophilic cellulitis as a result of onchocerciasis. Clin Exp Dermatol.1983; 8 :405 –408
Hurni MA, Gerbig AW, Braathen LR, Hunzinker T. Toxocariasis and Wells' syndrome: a causal relationship? Dermatology.1997; 195 :325 –328
Brehmer-Andersson E, Kaaman T, Skog E, Frithz A. The histopathogenesis of the flame figure in Wells' syndrome based on five cases. Acta Derm Venereol.1986; 66 :213 –219
Seckin D, Demirhan B. Drugs and Wells' syndrome: a possible causal relationship? Int J Dermatol.2001; 40 :133 –152
Coldiron BM, Robinson JK. Low-dose alternate-day prednisone for persistent Wells' syndrome. Arch Dermatol.1989; 125 :1625 –1626
Spigel GT, Winkelmann RK. Wells' syndrome: recurrent granulomatous dermatitis with eosinophilia. Arch Dermatol.1979; 115 :611 –613
Ferrier MC, Janin-Mercier A, Souteyrand P, et al. Eosinophilic cellulitis (Wells' syndrome): ultrastructural study of a case with circulating immune complexes. Dermatologica.1988; 176 :299 –304
Koh KJ, Warren L, Moore L, et al. Wells' syndrome following thiomersal-containing vaccinations. Australas J Dermatol.2003; 44 :199 –202
Moreno M, Luelmo J, Monteagudo M, Bella R, Casanovas A. Wells' syndrome related to tetanus vaccine. Int J Dermatol.1997; 36 :524 –525
Horn MJ, Katz DA, Bewtra C, Fusaro RM, Koh JK. Eosinophilic cellulitis (Wells' syndrome). Arch Dermatol.1985; 121 :836

Varrotti C, Tosti A, Gobbi M, Martinelli G, Patrizi A. Eosinophilic cellulitis: a new case. Dermatologica.1982; 164 :404 –406
Farrar CW, Guerin DM, Wilson NJE. Eosinophilic cellulitis associated with squamous cell carcinoma of the bronchus. Br J Dermatol.2001; 145 :667 –690
Murray D, Eady RAJ. Migratory erythema and eosinophilic cellulitis associated with nasopharyngeal carcinoma. J R Soc Med.1981; 74 :845 –847
Zachary CB, Wong E, Wilson-Jones E. Eosinophilic cellulitis [in French]. Ann Dermatol Venereol.1984; 111 :777 –778
Espana A, Sanz ML, Sola J, et al. Wells' syndrome (eosinophilic cellulitis): correlation between clinical activity, eosinophil levels, eosinophil cation protein and interleukin-5. Br J Dermatol.1999; 140 :127 –130
French LE, Shapiro M, Junkins-Hopkins JM, et al. Eosinophilic fasciitis and eosinophilic cellulitis in a patient with abnormal circulating clonal T cells: increased production of interleukin 5 and inhibition by interferon . J Am Acad Dermatol.2003; 49 :1170 –1174
Mould AW, Matthaei KI, Young IG, Foster PS. Relationship between interleukin-5 and eotaxin in regulating blood and tissue eosinophilia in mice. J Clin Invest.1997; 99 :1064 –1071
[Abstract/Free Full Text]
Marks R. Eosinophilic cellulitis: a response to treatment with dapsone: case report. Aust J Dermatol.1980; 21 :10 –12
Herr H, Jai-Kyoung K. Eosinophilic cellulitis (Wells' syndrome) successfully treated with low-dose cyclosporine. J Korean Med Sci.2001; 16 :664 –668
Husak R, Goerdt S, Orfanos CE. Interferon treatment of a patient with eosinophilic cellulitis and HIV infection. N Engl J Med.1997; 337 :641 –642
[Free Full Text]
Dirdil E, Honigsmann H, Tanew A. Wells' syndrome responsive to PUVA therapy. Br J Dermatol.1997; 137 :479 –481
Aroni K, Aivaliotis M, Liossi A, Davaris P. Eosinophilic cellulitis in a child successfully treated with cetirizine. Acta Derm Venereol.1999; 79 :332
Correia S, Garcia e Silva L. Childhood Wells syndrome [in Portuguese]. Med Cutan Ibero Lat Am.1988; 16 :221 –224
Garty BZ, Feinmesser M, David M, et al. Congenital Wells' syndrome. Pediatr Dermatol.1997; 14 :312 –315
Kuwahara RT, Randall MB, Eisner MG. Eosinophilic cellulitis in a newborn. Pediatr Dermatol.2001; 18 :89 –90
Davis MD, Brown AC, Blackston D, et al. Familial eosinophilic cellulitis, dysmorphic habitus, and mental retardation. J Am Acad Dermatol.1998; 38 :919 –928