Int Ophthalmol (2012) 32:361–367 DOI 10.1007/s10792-012-9574-9
ORIGINAL PAPER
Orbital and preseptal cellulitis: a 10-year survey of hospitalized patients in a tertiary eye hospital in Iran Abbas Bagheri • Mehdi Tavakoli • Maryam Aletaha • Hossein Salour Mahboubeh Ghaderpanah
•
Received: 4 May 2011 / Accepted: 20 April 2012 / Published online: 3 May 2012 Ó Springer Science+Business Media B.V. 2012
Abstract To evaluate the demographic and clinical features of orbital cellulitis (OC) and preseptal cellulitis (PC) in a tertiary eye hospital over a 10-year period. This is a retrospective study of files of patients defined as OC or PC admitted to Labbafinejad Hospital, Tehran, Iran, in which clinical specificities and results of treatment and complications were reviewed. A total of 93 records of patients with OC or PC were identified from 1997 to 2007. 42 % of the patients had orbital and 58 % had PC. OC was found to be twice as common in males but PC occurred equally in both sexes. The mean ages ± SD of patients with OC and PC were 27.4 ± 23.9 and 19.1 ± 23.3 years, respectively. 97.8 % of involvements were unilateral. The most common complaint of the patients was change in lid appearance and the most common season of involvement was spring in both groups. Sinusitis was the most common background in 53.8 % of OC and 24.1 % of PC patients, and the sinus most commonly involved was the ethmoid. Surgical intervention was needed in 48.7 % of orbital and 14.8 % of PC. The only complication seen in OC was external ophthalmoplegia in a case which needed operation. In our geographical region, orbital and PC are seen most frequently in young patients in spring and their most
common background is ethmoiditis; complications may be prevented by appropriate medical and/or surgical management. Keywords Orbital cellulitis Preseptal cellulitis Orbit Paranasal sinuses
Introduction Preseptal and orbital cellulitis (OC) are two major infectious disorders of the orbit and ocular adnexae. Preseptal cellulitis (PC) is infection of the soft tissues of the lids and periocular area located anterior to the orbital septum. If the infectious process extends beyond the septum, the term OC is used. These conditions, in particular OC, may lead to serious complications including blindness, intracranial extension of the infection, and death [1]. In addition, the disease is more prevalent in children and young individuals. We designed this retrospective study in order to have a better knowledge of the epidemiologic features and evaluation of etiologies, therapeutic results, and complications of preseptal and OC in our geographical region.
Methods A. Bagheri M. Tavakoli (&) M. Aletaha H. Salour M. Ghaderpanah Ophthalmic Research Center, Shahid Beheshti University of Medical Sciences, Tehran, Iran e-mail:
[email protected]
In this retrospective cross-sectional study, data were collected from the records of the patients admitted with the initial diagnosis of preseptal or OC to
123
362
Labbafinejad Hospital, Tehran, Iran, from January 1997 to December 2007. Routinely, all cases of OC and the severe cases of PC are admitted and undergo treatment, since the mild cases of PC are treated as outpatients and thus we could not access their files. The medical files were evaluated and the following data were extracted: age, gender, the involved eye, symptoms and signs at admission, visual acuity, restriction of ocular motions, funduscopic findings, history of trauma, skin wound, dental infection, previous surgeries, accompanying systemic disorders, medications taken before presentation, imaging [computed tomography (CT) or magnetic resonance imaging (MRI)] findings especially sinusitis and foreign body (in lids, globe or orbit), type and duration of medical treatment, type of surgery (if any), microbiologic findings, duration of admission, results of treatment, and complications during or after treatment. All patients had a CT scan and MRI, if needed, and differentiation between OC and PC was based on clinical findings as well as the imaging findings such as proptosis, intraorbital and/ or subperiosteal abscess, and loss of normal orbital fat shadow in OC but not in PC; however, reactive edema in the retroseptal space might exist in both diseases. The therapeutic results were considered excellent if the patient was treated successfully and without complications, good if complications occurred but consequently resolved, and poor or bad in the cases of persistent complications. Analyses of the data were performed with SPSS software using v2 test, t test, Fisher exact test, paired t test, Mann–Whitney and Pearson tests.
Int Ophthalmol (2012) 32:361–367
The mean age ± SD of the patients was 27.4 ± 23.9 years in the OC group, with a range of 6 months to 48 years and median of 15 years, and 19.1 ± 23.3 years in the PC group, with a range of 1 month to 82 years and median of 6 years. These findings show higher mean age in the OC group but this difference was not statistically significant (t test, p = 0.1). The age distribution in both groups skewed to the right (e.g. more prevalence of the disease among the lower ages) which was steeper in PC than OC (Fig. 2). We classified the patients according to age into toddler (up to 2 years of age), child (between 2 and 12 years of age) and adult (over 12 years of age). Selection of antibiotics in our ward, with some exceptions, had been according to this classification. When OC and PC were analyzed separately as in Fig. 3, we observed that PC is more common in younger and OC is more common in older patients. The frequency of the disease was similar in right and left eyes in both groups (Fisher exact test, p = 0.45) and in 97.8 % of the cases the involvement was unilateral. Only two cases (one case of PC and one case of OC) had bilateral presentation. The most common complaint in both groups was redness or swelling of the lids (Table 1) following by pain, especially in the OC group. Lid swelling and erythema were the most common findings on clinical examination in both groups. In the OC group the most common
Results In total, 93 records of patients with OC or PC were identified during a 10-year period, comprising 39 cases (42 %) of OC and 54 cases (58 %) of PC. 55 patients (59.1 %) were male and 38 patients (40.9 %) were female. The distribution of males and females in cellulitis indicates a twofold higher frequency of OC in men but PC had nearly equal frequency in each gender and the difference was not significant (v2 test, p = 0.2, Fig. 1). In addition, the ratio of prevalence of PC to OC was higher in women than in men, although not significantly (v2 test, p = 0.2).
123
Fig. 1 Gender distribution of OC and PC
Int Ophthalmol (2012) 32:361–367
restricted gaze was upward (79.5 %), followed by inward (69.2 %), outward (61.5 %), and downward (53.8 %). Visual acuity had been recorded quantitatively in 61 patients, including 35 patients with OC and 26 patients with PC, using the Snellen chart. In 32 patients the visual acuity had not been recorded because of low age in 31 patients and mental retardation in one patient. The mean LogMAR in the involved eye was 0.56 ± 0.84 in the OC group and 0.31 ± 0.57 in the PC group (t test, p = 0.02, Fig. 4). Both PC and OC occurred most commonly in spring (from April to June) and least commonly in summer (from July to September); in addition, in spring, OC was more common than PC but PC was more common in other seasons although the difference was not significant (Table 1). The most frequent etiology in both groups was sinusitis. Frequencies of other etiologies are summarized in Table 2. The ethmoid sinus was the most commonly affected sinus in both groups (46.4 % in OC and 19.4 % in PC) followed by the maxillary, frontal, and sphenoid sinuses respectively. In more detail, two of seven patients with a history of trauma had orbital fracture. Among the patients with a history of previous surgery, three patients had dental
363
Fig. 3 Frequency of OC and PC according to the defined age groups
extraction, two patients had orbital surgery, two patients had strabismus surgery, one patient had lid surgery, and the last patient had corneal transplantation. Six patients had dacryocystitis which was acute in three and equally distributed between the OC and
Fig. 2 Age distribution of patients according to type of cellulitis
123
364
Int Ophthalmol (2012) 32:361–367
Table 1 Demographic and clinical features of orbital and PC Orbital cellulitis
Preseptal cellulitis
p value
Mean age (years)
27.4 ± 23.9
19.1 ± 23.3
0.1a
Male (%)
66.7
53.7
0.2b
Right to left eye ratio
1.29
0.79
0.45c
Interval between presentation and referral to hospital (days)
3.9 ± 4
3.8 ± 5.3
0.95a
Lid redness and edema
89.7
100
0.03b
Periocular pain
61.5
31.5
0.004c
Proptosis
46.2
0.0
\0.01b
Reduced vision
20.5
0.0
0.001b
Ophthalmoplegia
84.6
3.8
\0.01b
Proptosis
46.2
1.9
\0.01b
Clinical abscess Ptosis
25.6 2.6
9.4 1.9
0.04b [0.99b
0.56 ± 0.84
0.31 ± 0.57
0.02b
Spring
41.02
29.6
Summer
12.82
22.2
Autumn
20.5
25.9
Sinusitis
Winter
25.6
22.2
Duration of hospitalization (days)
6.3 ± 3.8
4.5 ± 2.4
0.01a
Skin wound
Number of antibiotics
2.3 ± 0.7
2.0 ± 0.8
0.09d
Days of antibiotic use
5.4 ± 2.7
4.1 ± 2.0
0.01a
Symptoms (%)
Signs (%)
LogMAR visual acuity Seasonality (%)
Therapeutic results (%) Excellent Good
38.5 59
64.8 35.2
0.04b 0.02b
Bad
2.5
0.0
0.01b
a
t test
b
v2 test
c
Fisher exact test
d
Mann–Whitney test
PC groups. In 16 patients (17.2 %), no etiology was found. Four patients with OC yielded positive cultures including two cases of Staphylococcus aureus and two cases of a combination of S. aureus and betahemolytic streptococcus in the sample from abscess contents. On the other hand, in the PC group five patients were S. aureus positive and in one case a
123
Fig. 4 Difference between involved and non-involved eyes in each group in terms of LogMAR
Table 2 Prevalence of background etiologies in preseptal and OC Orbital (%)
Preseptal (%)
Total (%)
p value
21 (53.8)
13 (24.1)
34 (36.6) \0.01*
2 (5.1)
12 (22.2)
14 (15.1)
0.03*
Periocular surgery
4 (10.3)
5 (9.2)
9 (9.7)
[0.99
Trauma
4 (10.3)
3 (5.5)
7 (7.6)
0.46
Dacryocystitis
3 (7.7)
3 (5.5)
6 (6.4)
0.69
Foreign body Dental infection
2 (5.1) 1 (2.6)
1 (1.9) 2 (3.8)
3 (3.2) 3 (3.2)
[0.99 [0.99
Unknown causes
2 (5.1)
15 (27.8)
17 (18.2) [0.99
v2 test was used in analysis of all items except the cases identified with ‘‘*’’ in which the Fisher exact test was used
combination of pneumococci and Klebsiella was obtained from conjunctival swabbing. In other patients the cultures were negative in spite of efforts to obtain samples from conjunctiva, blood, throat, lacrimal sac contents, and abscess contents. The antibiotic regimen including type and doses in each patient had been selected according to the patients’ age, severity of involvement, and the results of consultation with the infectious disease specialist. The antibiotic types used in our series included cephalotin, gentamicin, amikacin,
Int Ophthalmol (2012) 32:361–367
cefazolin, ceftazidime, ceftriaxone, vancomycin, metronidazole, ciprofloxacin, ampicillin, clindamycin, and cloxacillin. The other details of antibiotic usage are shown in Table 1. The age of the patients had a direct relationship with the duration of antibiotic usage; older patients had a longer duration of antibiotics usage. (Pearson test, r = 0.36 and p = 0.03 for OC; r = 0.28 and p = 0.04 for PC). In the toddlers the most commonly used antibiotics included ceftazidime and cloxacillin in the OC group and ceftriaxone, ceftazidime, vancomycin, and gentamicin in the PC group. In the children, the most commonly used antibiotics were gentamicin, cephalotin, ceftriaxone, and vancomycin in OC and ceftriaxone, cephalotin, vancomycin, and gentamicin in PC. In the adults, the most commonly used antibiotics were vancomycin, ceftriaxone, ceftazidime, and metronidazole in OC and ceftazidime, cephalotin, and gentamicin in PC. Surgical intervention was performed in 29 % of all patients, 48.7 % of patients with OC and 14.8 % of patients with PC (Fisher exact test, p = 0.002). 15 % of surgeries were performed in the ophthalmology ward, 12.9 % in the ENT ward and 1.1 % in both wards. No patient needed neurosurgical intervention. The mean
365
age of the operated patients was 16.5 ± 3.5 years and the mean age of the non-operated patients was 24.6 ± 5.8 years; the difference was not significant (t test, p = 0.18). 20 % of the adults and 36.7 % of the children needed surgery whereas none of the toddler patients needed surgery (v2 test, p = 0.013). In the PC group, all patients improved without permanent complications. In the OC group, ophthalmoplegia and ptosis developed in one patient without improvement after resolution of inflammation, and thus surgical intervention was necessary (Fig. 5). Orbital or lid abscess formation developed in 46.2 % of OC patients and 1.9 % of PC patients (v2 test, p \ 0.001). The mean follow-up period was 38 months (range 3–84 months). During this period the mean LogMAR in the involved eye improved to 0.29 ± 0.41 in the PC group (t test, p = 0.64) and to 0.38 ± 0.69 in the OC group (t test, p = 0.03).
Discussion We retrospectively studied records of 93 patients admitted with the diagnosis of preseptal or OC over a
Fig. 5 Therapeutic results according to age groups
123
366
10-year period in our tertiary referral ophthalmology ward; 42 % of the patients were diagnosed as OC and the others were diagnosed as PC. Notably, many patients with a mild to moderate PC are treated as outpatients. Therefore such studies which are based on the review of the hospital records have some limitation in accurate comparison of orbital and PC and we should emphasize that all findings and comparisons in this survey relate to OC and severe PC (severe enough to be admitted to the hospital). Nevertheless, the incidence and prevalence of PC are generally considered to be greater than OC. The incidence of PC and OC was reported to be 84 and 16 % by Uzca´tegui et al. [2] and 72 and 28 % by Jackson and Baker [3]. In an epidemiological study of orbital infections in a 5-year period in the Philippines, 62 % of patients were defined as OC and the others were defined as PC. The ratio of male to female was 1.2 in the PC group and 2.2 in the OC group [4], very similar to our series in which this ratio was 1.15 and 2 in the PC and OC groups, respectively. The gender ratio of OC in a 15-year epidemiologic study in Saudi Arabia was estimated to be about 1.65 and in another study in Nigeria it was estimated to be about 2, although in a similar study from Spain no significant difference was reported between the two genders [5–7]. In addition, in a 10-year study in the US regarding OC in children the prevalence of OC among boys was 2.7 times more than girls [8]. This corresponds with the sex-related tendency of other serious childhood infections. The mean age of the patients in our study was 27.4 years in the OC group and 19.1 years in the PC group. The median and modal ages in each group were 15 years or less which demonstrates the higher prevalence of disorders among children, as shown in Fig. 3. The mean age in the studies in Saudi Arabia, Senegal, and Nigeria were 25.7, 18, and 13 years, respectively, similar to our results [5, 6, 9]. In the report by Nageswaran et al. [8] 73 % of OC patients suffered from proptosis, ophthalmoplegia, or both. These values in our study were 46.2 % for proptosis and 84.6 % for ophthalmoplegia. The most common presenting symptoms in Uy and Tuano [4] series were lid swelling and redness (respectively 100 and 34 %), quite similar to our results. In our records, both types of cellulitis were most common in spring and least common in summer, in contrast to Rodrı´guez Ferran et al. [7] report in which OC and PC were most commonly seen during winter
123
Int Ophthalmol (2012) 32:361–367
months. Jain and Rubin [10] also believe that cellulitis is more prevalent during winter months, corresponding with higher prevalence of sinusitis on these cold days. Other authors have reported that orbital and PC are seen more in the cold months of late fall to early spring, compatible with the peak of upper respiratory viral infections [11]. The higher prevalence of cellulitis in spring in our series may be related to the higher occurrences of allergies and aggravation of the background sinusitis in this season. The most common etiology of OC and PC in our series was sinusitis followed by trauma and periocular surgeries in the OC group and superficial wounds in the PC group. The most commonly involved sinus was the ethmoid sinus followed by the maxillary sinus. Similar results have been reported by several authors [4, 5, 8, 12, 13]. In one study conducted for evaluation of risk factors of orbital and PC in Pakistan, the patients were divided into two groups: group I included patients aged 6–16 years and group II included ages 17–40 years. In group I, the most common risk factor identified in 40 % of patients with PC was insect bite whereas the most common cause of OC in 50 % of the patients was trauma. In group II, trauma was the leading cause in 50 % of patients with PC and sinusitis was a common cause in 18.1 % with those with OC [14]. Some other rare etiologies in our patients like strabismus surgery and dental infection have been previously reported [15, 16]. In our series, we encountered a case of OC following corneal transplantation. To our knowledge, such a case has not been reported in the literature and this association may be an incidental finding. The selection of antibiotics in our patients is very similar to other reports [4, 8, 17, 18]. The coverage of common organisms associated with acute sinusitis and other common etiologies should be considered in the empirical medical management of preseptal and OC. However, in cases of subperiosteal abscess formation, lack of appropriate response to medical treatment, intracranial involvement or threatened visual acuity, surgical intervention is indicated. In our series, 29 % of patients needed surgery during the course of treatment whereas 71 % improved only with medical management. The OC patients needed surgery more frequently than the PC patients. In the report by Uy and Tuano [4], 36 % of PC patients and 63 % of OC patients needed surgical treatment including incision and drainage of the lid or orbital abscesses, sinus surgery, etc.
Int Ophthalmol (2012) 32:361–367
The limitations of our study included the retrospective design in which the probability of missing some data is inevitable, as well as unavailability of the records of the patients with PC who had been treated as outpatients. Acknowledgments The authors would like to thank M. Yaseri, PhD for his major role in statistical analysis and M. Hadi Tavakoli, MSc and R. Beheshti Zavareh, PhD for their assistance in revising and English editing the manuscript.
References 1. Wulk AE, Edmonson BC (2008) Orbital infections. In: Tasman W, Jaeger EA (eds) Duane’s clinical ophthalmology, revised edition. vol 2, Chap 34. Williams & Wilkins, Lippincott, pp 1967–1994 2. Uzca´tegui N, Warman R, Smith A, Howard CW (1998) Clinical practice guidelines for the management of orbital cellulitis. J Pediatr Ophthalmol Strabismus 35(2):73–79 (quiz 110-1) 3. Jackson K, Baker SR (1987) Periorbital cellulitis. Head Neck Surg 9(4):227–234 4. Uy HS, Tuano PM (2007) Preseptal and orbital cellulitis in a developing country. Orbit 26(1):33–37 5. Chaudhry IA, Shamsi FA, Elzaridi E, Al-Rashed W, AlAmri A, Al-Anezi F, Arat YO, Holck DE (2007) Outcome of treated orbital cellulitis in a tertiary eye care center in the Middle East. Ophthalmology 114(2):345–354 6. Bekibele CO, Onabanjo OA (2003) Orbital cellulitis: a review of 21 cases from Ibadan, Nigeria. Int J Clin Pract 57(1):14–16
367 7. Rodrı´guez Ferran L, Puigarnau Vallhonrat R, Fasheh Youssef W, Ribo´ Aristaza´bal J, Luaces Cubells C, Pou Ferna´ndez J (2000) Orbital and periorbital cellulitis. Review of 107 cases. An Esp Pediatr 53(6):567–572 8. Nageswaran S, Woods CR, Benjamin DK Jr, Givner LB, Shetty AK (2006) Orbital cellulitis in children. Pediatr Infect Dis J 25(8):695–699 9. Wane AM, Ba EA, Ndoye-Roth PA, Kameni A, Demedeiros ME, Dieng M, Ndiaye MR, Ndiaye PA, Ben Nasr S, Wade A (2005) Senegalese experience of orbital cellulitis. J Fr Ophtalmol 28(10):1089–1094 10. Jain A, Rubin PA (2001) Orbital cellulitis in children. Int Ophthalmol Clin 41(4):71–86 11. Oxford LE, McClay J (2005) Complications of acute sinusitis in children. Otolaryngol Head Neck Surg 133(1):32–37 12. Liu IT, Kao SC, Wang AG, Tsai CC, Liang CK, Hsu WM (2006) Preseptal and orbital cellulitis: a 10-year review of hospitalized patients. J Chin Med Assoc 69(9):415–422 13. Yang M, Quah BL, Seah LL, Looi A (2009) Orbital cellulitis in children-medical treatment versus surgical management. Orbit 28(2–3):124–136 14. Babar TF, Zaman M, Khan MN, Khan MD (2009) Risk factors of preseptal and orbital cellulitis. J Coll Physicians Surg Pak 19(1):39–42 15. Wilson ME, Paul TO (1987) Orbital cellulitis following strabismus surgery. Ophthalmic Surg 18(2):92–94 16. van der Borden J Jr, Steinmetz PA, Abraham-Inpijn L, Visser CE, Westerbeek HA (1989) Orbital cellulitis after a dental infection. Ned Tijdschr Tandheelkd 96(12):561–563 [Article in Dutch] 17. Starkey CR, Steele RW (2001) Medical management of orbital cellulitis. Pediatr Infect Dis J 20(10):1002–1005 18. Noe¨l LP, Clarke WN, MacDonald N (1990) Clinical management of orbital cellulites in children. Can J Ophthalmol 25(1):11–16
123
