Abstract
Background
Injury to the medial rectus (MR) is a potentially devastating complication of orbital and sinus surgery. Precise knowledge of the MR relative to the lamina papyracea (LP) is important during endoscopic surgery for both Graves’ ophthalmopathy and inflammatory disease. The objective of this study is to determine the location of the MR in relation to easily identified and frequently encountered intranasal landmarks in patients with and without Graves’ disease.
Methods
High-resolution computed tomography scans were analyzed in 100 controls and 63 patients with Graves’ disease. The MR position was recorded relative to the maxillary sinus ostium (MSO), anterior ethmoid artery (AEA), and posterior ethmoid artery (PEA)/horizontal basal lamella (BL). Clinically relevant variables recorded at each level included the Keros stage, AEA position, MR height, and distance of the MR to orbital floor, skull base, and LP.
Results
The mean distances between the MR and LP were statistically different for both groups. Controls at the MSO, AEA, and PEA/BL were 2.92, 1.69, and 1.06 mm; for Graves’ patients measurements at these sites were 2.12, 1.20, and 0.029 mm. When comparing the two groups, each of these distances were statistically significant ( p < 0.02). There was no difference in ethmoid cavity width ( p > 0.05) between controls (9.66 mm) and Graves’ patients (9.70 mm). Sex, age, and skull base depth were not statistically significant factors.
Conclusion
This study illustrates the position of the MR from the perspective of an endoscopic surgeon utilizing fixed intranasal landmarks. Knowledge of the position of MR is critical to safely perform decompression surgery and when operating adjacent to the LP during endoscopic surgery.
1
Introduction
Medial rectus (MR) injury is a rare but potentially devastating complication of endoscopic sinus surgery (ESS). The proximity of the MR to the ethmoid sinus and lamina papyracea makes it the most commonly injured extraocular muscle in ESS . Injury may occur via direct trauma, neurovascular interruption, or secondary scarring and fibrosis to adjacent structures . Even partial muscle injury may lead to permanent diplopia due to reduced or absent globe adduction. Despite surgical and medical interventions, the prognosis for complete return of MR function remains disappointing, and thus prevention of injury remains the primary goal .
Graves’ disease is an autoimmune condition that may lead to disfiguring exophthalmos, dysthyroid optic neuropathy, or exposure keratopathy requiring surgical decompression. Radiographic findings found in Graves’ ophthalmopathy include enlargement of extraocular muscles, intra- and extraconal fat expansion, bowing of the medial orbital wall, and optic nerve compression . The disease is usually bilateral but often asymmetric. Decompression causes expansion of the orbital cavity volume, allowing retrograde globe placement, and may be performed via an open or endoscopic approach. Endoscopic decompression involves an endonasal approach, with removal of the lamina papyracea (LP) followed by periorbital incisions to produce orbital fat prolapse. Enlargement of the MR and lateral bowing of the LP in these patients would theoretically place the orbital contents and MR at increased risk for injury during ESS and endoscopic orbital decompressions.
The primary objective of this study is to describe the position of the MR relative to the LP from the perspective of an endoscopic surgeon using commonly encountered and easily identifiable intranasal surgical landmarks. The second objective is to determine the differences between patients with and without Graves’ disease in regard to these dimensions. Additional analysis was performed to identify any effects of age, sex, Keros classification, and presence of a low-lying anterior ethmoid artery (AEA) upon these relationships. We additionally introduce a technical modification to improve the safety of endoscopic orbital decompression based on our findings. To the best of our knowledge, no previously published studies have described these findings.
2
Methods
High-resolution computed tomography (CT) scans were analyzed in 100 controls and 63 patients with Graves’ disease. This study was approved by the institutional review boards of the University of California, Los Angeles and the University of Colorado. CT imaging was performed of the paranasal sinuses without contrast on a 64 detector row scanner. Images were acquired helically in a plane parallel to the infraorbitomeatal line, with 0.6 mm detector width, 120 kVp, and 200 mAs with attenuation-based tube current modulation. Measurements were performed on the picture archiving and communication system (PACS) planning station utilizing the ruler tool by two independent reviewers, with the two measurements averaged to minimize interrater variability.
For the control group, 100 CT scans were reviewed from 100 unique patients from a tertiary academic rhinology practice. Measurements were taken from both the right and left sides, for 200 total measurements. Inclusion criteria included: 1) adults > 18 years of age and 2) minimal to absent sinus disease. Exclusion criteria included: 1) prior sinonasal surgery; 2) history of immunodeficiency, cystic fibrosis, or primary ciliary dyskinesia; 3) space occupying or expansile lesions (i.e., tumors or fungal sinusitis); 4) severe sinus disease on the CT scan making measurement of intranasal landmarks difficult; 5) evidence of craniofacial trauma on CT; and 6) studies of poor technical quality.
For the group with Graves’ ophthalmopathy, 63 preoperative scans were evaluated from 63 unique patients. Measurements were taken from both the right and left sides, for 126 total measurements. The diagnosis of Graves’ ophthalmopathy was made by an oculoplastic surgeon prior to the CT scan. Inclusion and exclusion criteria were similar to the control group, with the exception of including patients with a diagnosis of Graves’ disease. All patients in this group subsequently underwent an endoscopic decompression for thyroid eye disease.
For all patients, the position of the MR was recorded at three separate coronal planes—relative to the locations of the maxillary sinus ostium (MSO), AEA, and posterior ethmoid artery (PEA)/horizontal segment of the basal lamella (BL) ( Fig. 1 ). The latter two landmarks can be seen as pyramidal-shaped protrusions with a wider base originating from the superomedial orbit and traveling medially toward the ethmoid roof (“nipple sign”). These anatomic landmarks were selected because they are readily identifiable on preoperative coronal CT scans, as well as may be approximated intraoperatively. Clinically relevant variables recorded for analysis included: Keros stage based on olfactory fossa depth (type 1: 0–3 mm, type 2: 4–7 mm, type 3: > 7 mm) at the coronal cut demonstrating the lowest point of the olfactory fossa , AEA position (inferior to or at level of skull base), MR height, ethmoid cavity width (as measured from middle turbinate to lamina papyracea) at the level of the AEA, and distance of the MR to orbital floor (OF), skull base (SB), and lamina papyracea (LP). All measurements were parallel to the sagittal or coronal planes to reflect true superior-inferior and medial-lateral distances, respectively.
For a given measurement, means were compared using a repeated analysis of variance (ANOVA) model with group (control or Graves’) site, sex, Keros stage, AEA position, and side (left versus right) as the predictive factors. Calculations were carried out using SAS 9.3 (SAS Inc., Cary, NC, USA). A significance level of 0.05 was used for all analyses.
2
Methods
High-resolution computed tomography (CT) scans were analyzed in 100 controls and 63 patients with Graves’ disease. This study was approved by the institutional review boards of the University of California, Los Angeles and the University of Colorado. CT imaging was performed of the paranasal sinuses without contrast on a 64 detector row scanner. Images were acquired helically in a plane parallel to the infraorbitomeatal line, with 0.6 mm detector width, 120 kVp, and 200 mAs with attenuation-based tube current modulation. Measurements were performed on the picture archiving and communication system (PACS) planning station utilizing the ruler tool by two independent reviewers, with the two measurements averaged to minimize interrater variability.
For the control group, 100 CT scans were reviewed from 100 unique patients from a tertiary academic rhinology practice. Measurements were taken from both the right and left sides, for 200 total measurements. Inclusion criteria included: 1) adults > 18 years of age and 2) minimal to absent sinus disease. Exclusion criteria included: 1) prior sinonasal surgery; 2) history of immunodeficiency, cystic fibrosis, or primary ciliary dyskinesia; 3) space occupying or expansile lesions (i.e., tumors or fungal sinusitis); 4) severe sinus disease on the CT scan making measurement of intranasal landmarks difficult; 5) evidence of craniofacial trauma on CT; and 6) studies of poor technical quality.
For the group with Graves’ ophthalmopathy, 63 preoperative scans were evaluated from 63 unique patients. Measurements were taken from both the right and left sides, for 126 total measurements. The diagnosis of Graves’ ophthalmopathy was made by an oculoplastic surgeon prior to the CT scan. Inclusion and exclusion criteria were similar to the control group, with the exception of including patients with a diagnosis of Graves’ disease. All patients in this group subsequently underwent an endoscopic decompression for thyroid eye disease.
For all patients, the position of the MR was recorded at three separate coronal planes—relative to the locations of the maxillary sinus ostium (MSO), AEA, and posterior ethmoid artery (PEA)/horizontal segment of the basal lamella (BL) ( Fig. 1 ). The latter two landmarks can be seen as pyramidal-shaped protrusions with a wider base originating from the superomedial orbit and traveling medially toward the ethmoid roof (“nipple sign”). These anatomic landmarks were selected because they are readily identifiable on preoperative coronal CT scans, as well as may be approximated intraoperatively. Clinically relevant variables recorded for analysis included: Keros stage based on olfactory fossa depth (type 1: 0–3 mm, type 2: 4–7 mm, type 3: > 7 mm) at the coronal cut demonstrating the lowest point of the olfactory fossa , AEA position (inferior to or at level of skull base), MR height, ethmoid cavity width (as measured from middle turbinate to lamina papyracea) at the level of the AEA, and distance of the MR to orbital floor (OF), skull base (SB), and lamina papyracea (LP). All measurements were parallel to the sagittal or coronal planes to reflect true superior-inferior and medial-lateral distances, respectively.
