We read the recent correspondence from Dr Carifi and associates, and we thank them for their interest in our research as well as their comments. We would like to address their points below.

Dr Carifi makes the point that, given the spectrum of severity in orbital infections, a 3-tier grouping of patients might be beneficial in evaluating the utility of the initial ophthalmic examination in predicting surgical intervention in each group. In our view, it is likely that the predictive value of the initial ophthalmic examination would remain consistent across patients grouped by Chandler classification. Evidence of this can be seen in Table 3 in our publication, which uses a logistic regression model where each risk factor is reported independently of other listed risk factors. In this table, presence of an abscess is a clear risk factor for surgical intervention, but proptosis, elevated intraocular pressure (IOP), age greater than 9 years, and extraocular motility (EOM) restriction each represents an independent risk factor. This means that, abscess or not, the aforementioned examination findings are risk factors for surgery in and of themselves.

Next, Dr Carifi makes the point that the degree of proptosis, not the presence of proptosis, may be a more useful risk factor. He also notes skepticism in the use of palpating the globe to assess “IOP as firm or soft.” We agree on both of these points, and note that we have addressed them in the discussion section of our article. In practicality, Hertel’s exophthalmometry and tono-applanation in young children in acute pain in the emergency department is not frequently achievable or reliable. In these cases, grading of proptosis and intraocular pressure by the consulting ophthalmology service using visual and tactile methods is the most accurate as well as reflective of the ophthalmic examination in actual clinical practice.

The next question is regarding the details of our antibiotic treatment algorithm. In our institution, treatment always starts with broad-spectrum antibiotics, typically ampicillin/sulbactam or clindamycin, each with the addition of ceftriaxone. If methicillin-resistant Staphylococcus aureus is suspected, vancomycin is added. The antibiotic regimen is then narrowed based on culture results.

Lastly, Dr Carifi brings up the question of surgical approach, specifically when to use an external-only, endoscopic-only, or combined endoscopic and external approach. This is a very good question, and in our experience a combined approach frequently yields fewer postoperative complications. In fact, our group has investigated this question and will be publishing an article in the forthcoming literature to address surgical approaches specifically.

In sum, we thank Dr Carifi and associates for their close reading of our article and we appreciate the opportunity to address their thoughtful questions.