Figure. Angiograms, with orbitovascular anatomy shown below. A, Posteroanterior (left) and lateral (right) views of the left orbit showing the presence of dye in the lacrimal artery. B, Anteroposterior (left) and lateral (right) views of the right orbit showing the absence of dye in the lacrimal artery; there is partial filling at the beginning of the lacrimal artery, which creates an angiographic stub where the dye meets blood from collateral circulation without dye. See videos of the full angiograms.
Marr BP, Hung C, Gobin YP, Dunkel IJ, Brodie SE, Abramson DH. Success of Intra-arterial Chemotherapy (Chemosurgery) for RetinoblastomaEffect of Orbitovascular Anatomy. Arch Ophthalmol. 2012;130(2):180-185. doi:10.1001/archophthalmol.2011.386
Author Affiliations: Ophthalmic Oncology Service (Drs Marr, Hung, and Abramson) and Department of Pediatrics (Dr Dunkel), Memorial Sloan-Kettering Cancer Center, Interventional Neuroradiology, Departments of Radiology, Neurosurgery, and Neurology, Weill Cornell Medical College (Dr Gobin), and Department of Ophthalmology, Mount Sinai School of Medicine (Dr Brodie), New York, New York.
Objectives To review results of orbital angiography performed during intra-arterial chemotherapy (chemosurgery) for treatment of retinoblastoma to assess the association of angiographic variability in orbitovascular anatomy with tumor response and outcomes.
Methods Medical records and 64 orbital angiograms were reviewed for 56 pediatric patients with retinoblastoma undergoing chemosurgery using a combination of melphalan hydrochloride, topotecan hydrochloride, or carboplatin. The major orbital arteries and capillary blush patterns were graded, and tumor response and recurrence were compared using the log-rank and Fisher exact tests.
Results Statistically significant variables for tumor response were lacrimal artery prominence (P = .001), previous treatment (P = .003), and lacrimal blush (P = .004). The only statistically significant variable for vitreous seed response was ciliary body blush (P = .03). Statistically significant variables influencing time to recurrence and time to enucleation were choroidal blush absence (P = .01) and lacrimal artery presence (P = .03), respectively.
Conclusions The success of intra-arterial chemotherapy is dependent on delivery of drug to the target tumor within the eye via the ophthalmic artery. Because of the small volume of drug used (0.50-1.25 mL per treatment) and the selectivity of catheterization, variables affecting orbital blood flow greatly influence drug delivery and the success of chemosurgery.
Four years ago, our group began treating retinoblastoma using a novel technique of infusing small doses (but locally very high concentrations) of chemotherapy via the ophthalmic artery (termed chemosurgery), and the results have been described previously.1,2 Early on, a notable difference in salvaging these eyes was dependent on whether prior therapy had been given. Ninety percent of eyes for which chemosurgery was the initial treatment modality were salvaged.3 In contrast, 50% to 66% of eyes that progressed after conventional multiagent systemic chemotherapy, focal treatments, and then radiatotherapy were salvaged.4 However, with more experience using chemosurgery, it was noted that some patients had a better response than others and we wondered if the basic anatomy of the individual orbit might be relevant. The greatest flow from the ophthalmic artery was not to the eye in most patients but to the supratrochlear, dorsonasal, and mediopalpebral artery complex, and there was considerable variation in the normal anatomy of orbital vessels (as Hayreh5 and Hayreh and Dass6,7 previously emphasized).
We suspected that this variability in orbitovascular anatomy affects the clinical response to chemosurgery. Therefore, we reviewed results of orbital angiography performed during chemosurgery for treatment of retinoblastoma. To date, pediatric orbital angiography in this patient population has not been described. We describe our evaluation of 64 initial pediatric orbital angiograms in 56 patients with retinoblastoma undergoing chemosurgery and examine the association of angiographic variability in orbitovascular anatomy with tumor response and outcomes.
Internal review board approval was obtained for review of medical records for patients with retinoblastoma undergoing chemosurgery between May 30, 2006, and August 11, 2009. This technique was offered as an alternative treatment in lieu of enucleation. Two of us (B.P.M. and C.H.) in a masked fashion graded 64 initial orbital angiograms from 56 patients. One of us (Y.P.G.) had performed all angiography and intra-arterial procedures in a standard manner. Prechemotherapy injection angiograms were compared with postchemotherapy injection angiograms to confirm that no variations occurred during injections and that the angiograms were consistent before and after drug delivery. The major orbital arteries were graded as present, absent, or prominent on angiograms (Figure, A). Observed were the supratrochlear, anteroethmoid, posteroethmoid, middle meningeal, supraorbital, muscular, lacrimal, posterociliary, and centroretinal arteries. Clinical photographs, drawings, and notes were evaluated for tumor response and vitreous seed response. Grading was based on the tumor volume or vitreous seed volume seen clinically at the beginning of chemosurgery and at the last follow-up examination. Low response corresponded to a 0% to 33% reduction in the initial clinical tumor volume or vitreous seed volume, medium as a 34% to 66% reduction, and high as at least a 67% reduction. Capillary filling patterns seen on angiograms were graded as present, absent, or prominent (Figure, B). Patient data were reviewed, and angiographic findings were compared between groups of patients who had and had not had tumor response, vitreous seed response, tumor recurrence, and enucleation. The association of previous (non–intra-arterial) treatment and vessel and blush data with disease response (low vs medium or high) was examined using the Fisher exact test. The association of previous treatment and vessel and blush data with time to recurrence or time to enucleation was examined using the log-rank test.
Patient characteristics are listed in Table 1. The mean follow-up time was 7.5 months (range, 1-37 months), which represents the interval used for grading of disease response. Outcome results of tumor response, vitreous seed response, tumor recurrence, and enucleation are listed in Table 2. Forty eyes had undergone previous treatment, of which 31 eyes received intravenous chemotherapy, 7 eyes received external beam radiotherapy (6 eyes with both intravenous chemotherapy and external beam radiotherapy), and 7 eyes received only plaque (3 eyes) or focal laser or cryotherapy (4 eyes) treatments. Statistically significant variables for tumor response were lacrimal artery prominence (P = .001), previous treatment (P = .003) (in which eyes without prior treatment had a greater response rate), and lacrimal blush (P = .004). Eyes with lacrimal artery prominence and lacrimal blush showed better tumor response and vitreous seed response than eyes without these angiographic findings. The only statistically significant variable for vitreous seed response was ciliary body blush (P = .03) (Table 3). Statistically significant variables influencing time to recurrence and time to enucleation were choroidal blush absence (P = .01) and lacrimal artery presence (P = .03), respectively (Table 4).
Experience with intra-arterial chemotherapy for the management of retinoblastoma has been described.2 In this study, we examine the association of angiographic blood flow patterns with clinical response within the eye. Hayreh5 and Hayreh and Dass6,7 previously detailed the orbitovascular anatomy in 386 adult cadaver eyes, emphasizing that no 2 orbits were the same and that they differed on the 2 sides of the same cadaver. Therefore, a microcatheter in the ophthalmic artery may deliver different proportions of drug even if it seems that the microcatheter is in the “same place.” For example, some eyes are supplied from a branch off the middle meningeal artery, which is a branch of the external and not internal carotid artery. In some eyes, the greatest flow from the ophthalmic artery is not to the eye at all but rather to the supratrochlear, dorsonasal, and mediopalpebral artery complex, which supplies the nose, eyelids, and inner forehead. However, anatomy does not predict blood flow, especially when a small catheter is placed into the ophthalmic artery because the catheter may alter the normal flow patterns.
The microcatheter is usually placed at the ostium of the ophthalmic artery for selective angiography and for chemosurgery. This microcatheter measures 0.45 mm in diameter, so its bare presence at the ostium of the ophthalmic artery may decrease blood flow through the ophthalmic artery, as well as distal perfusion pressure. This facilitates the compensatory redirection of collateral vascularization from the external carotid artery to orbital and ocular branches that will not be filled with the chemotherapy drugs. The ophthalmic artery provides the main arterial flow to the orbit from the internal carotid artery. The middle meningeal artery communicates with the lacrimal artery through the supero-orbital fissure via the recurrent meningeal artery.8,9 A rare anatomical variant exists in 4% to 6% of adult orbits,5,9 where the ophthalmic artery arises from the middle meningeal artery and not directly from the internal carotid artery. This was seen in one of our patients. Other anastomoses with the external carotid artery exist within the orbit through the superficial temporal artery and facial artery. The effect of these communications on ocular blood flow during ophthalmic artery infusion is largely unknown. The effect of anatomical variation of orbital vessels on the success of chemosurgery is also unknown. We attempted to identify angiographic patterns in orbitovascular anatomy that were predictive of success by observing tumor response and vitreous seed response in eyes with tumor recurrence and in eyes that required enucleation.
Variables affecting tumor response were lacrimal artery prominence, previous treatment, and lacrimal blush. Because tumor response was graded based on reduction in tumor volume, eyes that had received previous treatment had less tumor volume and subsequently less volume reduction. Hence, we believe that the observed association of previous treatment with poorer response is owing to the smaller size of previously treated tumors. Lacrimal artery presence or prominence and lacrimal blush were statistically significantly associated with tumor response. This may result from communication of the lacrimal artery with the middle meningeal artery. In 63% to 90% of lacrimal arteries studied, muscular branches arose from the lacrimal artery, and posterociliary arteries arose from it in 5% of lacrimal arteries studied.9- 11 Such angiographic findings may indicate dilution of drug to the eye via this communication. Vitreous seed response was affected by ciliary body blush presence or prominence. This suggests that vitreous drug levels may be influenced by ciliary body blood flow. Choroidal blush absence significantly affected time to recurrence, and lacrimal artery presence affected time to enucleation in our series.
This study represents one of the largest detailed examinations of pediatric orbital angiography. The study is limited by its short mean follow-up time relative to outcomes. We acknowledge the limitations of clinical observations in interpreting angiograms, tumor response, and vitreous seed response. This study highlights the anatomical variation in pediatric orbital vasculature and the dynamic nature of blood flow within the pediatric orbit, the latter of which is affected by the presence of the microcatheter. Angiographic patterns in orbitovascular anatomy, including ciliary body blush, choroidal blush, and lacrimal artery visualization, can give insight to the response to chemosurgery.
The success of chemotherapy is dependent on delivery of drug to the target tumor in the eye via the ophthalmic artery. Because of the small volume of drug used (0.50-1.25 mL per treatment) and the selectivity of catheterization, variables affecting orbital blood flow greatly influence drug delivery and the success of chemosurgery.
Correspondence: Brian P. Marr, MD, Ophthalmic Oncology Service, Memorial Sloan-Kettering Cancer Center, 70 E 66th St, New York, NY 10065.
Submitted for Publication: June 18, 2010; final revision received July 22, 2011; accepted August 1, 2011.
Financial Disclosure: None reported.
Funding/Support: This study was supported in part by a grant from the Fund for Ophthalmic Knowledge, Inc (Dr Abramson).
Additional Contributions: Elyn Renee Riedel, PhD, performed statistical work to the study.