Sinus-related orbital involvement and its sequelae have been classified into five stages by Chandler et al.
, with stage I corresponding to inflammatory oedema and stage V to cavernous sinus thrombophlebitis. In particular, the orbital complications of rhinosinusitis can be defined as peri-orbital cellulitis, orbital cellulitis, sub-periosteal abscess and orbital abscess depending on the site and entity of the inflammation. Peri-orbital cellulitis refers to the infection of tissues anterior to the orbital septum whereas, in the case of orbital cellulitis or abscess, the infection involves the posterior tissues; a sub-periosteal abscess is a pool of purulent material between the periorbita and the bony orbital wall under the periosteum
. Given the inflammatory involvement of the orbital contents posterior to the orbital septum, our patient belonged to stage II (orbital cellulitis)
The pathogens mainly involved are those generally responsible for acute rhinosinusitis, such as Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, Staphylococcus aureus, group A streptococcus, and upper respiratory tract anaerobes
[12, 13]. However, the introduction of the heptavalent pneumococcal vaccine and immunisation against Haemophilus influenzae type B have been associated with a decrease in the occurrence of invasive diseases such as peri-orbital and orbital cellulitis, and possibly a change in their microbiology
Although uncommon (the reported incidence in specialised tertiary centres is 0.3-1.3 cases per month)
, the disease processes may lead to serious complications, including vision loss and a predisposition for life-threatening events such as cavernous sinus thrombophlebitis or other intra-cranial complications
[3, 4, 6, 7]. The incidence of associated complications was especially high in the pre-antibiotic era, with 17% mortality due to meningitis and a 20% rate of blindness
, the rates of vision loss and overall mortality are now respectively 11% and 2.5%
. In addition, there may be less troublesome but sometimes long-lasting functional and aesthetic sequelae, such as residual asymmetrical eyelid opening, impaired ocular motility, and eyelid inflammation.
In our case, although a clinical improvement was documented a few days after beginning adequate intravenous broad spectrum antibiotic therapy according to the guidelines
, complete resolution of the clinical signs and symptoms of acute inflammation such as eyelid redness and swelling and orbital pain was only observed after 40 days, and complete clinical recovery with normally symmetrical eyelid opening was not documented until about seven months after disease onset. To the best of our knowledge, no similar cases have been reported in the literature.
There is no easy explanation for the prolonged course. In particular, conservative medical therapy (the first-line treatment for patients in Chandler stage II)
 was administered in accordance with the guidelines
[8, 18], and no unfavourable clinical or anatomical situations coexisted. As a matter of fact, A CT scan failed to detect any condition that may have been responsible for impaired sinusal drainage or the persistence of orbital and peri-orbital inflammation, such as ethmoidal or orbital mucoceles or lamina papyracea erosion. Furthermore, no other risk factors were detected as the patient had never undergone nasosinusal surgery and immunodeficiency was excluded.
This long-term evolution resulted in discomfort for the patient and great concern for the parents and clinicians. However, the long-term serial clinical and instrumental follow-up with the collection and comparison of detailed iconography allowed us to assess the slow but progressive clinical improvement, and exclude the persistence or recurrence of orbital inflammation or the development of such subtle and often asymptomatic sequelae as mucoceles. The long-lasting clinical course initially led us to plan magnetic resonance imaging in order to rule out any orbital or ethmoidal mucocele, but the albeit slow progressive clinical improvement made this unnecessary. This approach seemed to be appropriate as no recurrences have been detected during the currently 25 months of follow-up.
Sinus-related orbital infections are well represented in the international scientific literature
[1, 2, 5, 8, 11] but, to the best of our knowledge, the long-lasting course of pediatric peri-orbital and orbital cellulitis has not been previously assessed and there is no other detailed iconographic description of its temporal evolution.
In conclusion, we would like to emphasise the fact that the management of children with suspected sinus-related orbital infection is complex and requires a multidisciplinary approach involving pediatricians, otolaryngologists, ophthalmologists and radiologists. This not only allows a prompt diagnosis and adequate treatment, but also makes it possible to follow up and support children and their families during convalescence, and to detect any recurrence or related complications in a timely manner.