Original Investigation
Imaging Findings of Recurrent Tumors After Orbital Exenteration and Free Flap Reconstruction Paul S. Lee, M.D.*, Peter Sedrak, M.D.*, Nandita Guha-Thakurta, M.D.†, Edward I. Chang, M.D.‡, Lawrence E. Ginsberg, M.D.†, Bita Esmaeli, M.D.‡§, and James Matthew Debnam, M.D.† *Section of Neuroradiology, Department of Diagnostic Radiology, The University of Texas at Houston, Houston, Texas, U.S.A.; and †Section of Neuroradiology, Department of Diagnostic Radiology, ‡Department of Plastic Surgery, and §Orbital Oncology and Oculoplastic Surgery Program, The University of Texas MD Anderson Cancer Center, Houston, Texas, U.S.A.
Purpose: Tumors that recur following orbital exenteration may not be evident on clinical examination, highlighting the need for imaging surveillance. The goal of this study was to report the imaging characteristics of recurrent tumors following orbital exenteration and free flap reconstruction. Methods: The authors retrospectively reviewed the records of 48 patients who underwent orbital exenteration for the treatment of orbital malignancy and identified 17 recurrent tumors in 17 patients. The lesions were assessed for the presence of a soft tissue mass, imaging characteristics, and fluorodeoxyglucose avidity. Results: The recurrent tumors were detected 1 month to 6 years 10 months (median, 1 year 3 month) after orbital exenteration. On both CT and MRI, all 17 lesions were soft tissue masses at presentation. On CT, the lesions demonstrated heterogeneous to homogeneous to centrally necrotic enhancement; on MRI, the lesions were T1 hypointense to isointense and T2 hypointense to hyperintense. Twelve of the 15 recurrent tumors with available preoperative imaging had an enhancing appearance similar to that of the original tumor. Thirteen of the 17 recurrent tumors were at the margin of a flap placed for reconstruction; the other 4 lesions were remote from the operative site. Conclusion: Recurrent tumors following orbital exenteration and free flap reconstruction demonstrate a wide range of imaging appearances but most often appear as a soft tissue masses often similar in appearance to the primary tumor and arising near the flap margin. Awareness of the imaging features of recurrent disease is important because failure to diagnose recurrence can delay appropriate treatment. (Ophthal Plast Reconstr Surg. 2014;30:315–321)
O
rbital tumors arise from the structures within the orbit and periorbital region. Primary orbital tumors include lymphoproliferative tumors, mesenchymal tumors, and tumors of the lacrimal gland, including adenoid cystic carcinoma. Tumors Accepted for publication November 23, 2013. Supported by The University of Texas MD Anderson Cancer Center is supported, in part, by the National Institutes of Health through Cancer Center Support Grant CA016672. The authors have no conflicts of interest to disclose. Address correspondence and reprint requests to James Matthew Debnam, M.D., The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1482, Houston, TX 77030. E-mail: matthew.debnam@ mdanderson.org DOI: 10.1097/IOP.0000000000000100
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
arising from the paranasal sinuses can secondarily involve the orbit.1 Other primary tumors can also metastasize to the orbit. Treatment of aggressive orbital tumors may include orbital exenteration, a radical procedure that involves the removal of the soft tissue contents of the orbit, including the globe, extraocular muscle and optic nerve, eyelids, and adjacent skin. Orbital exenteration is usually reserved for aggressive orbital tumors or locally advanced sinonasal malignancies that are potentially fatal and relentlessly progressive.2–6 A variety of types of flaps and grafts can be used for reconstruction following exenteration.4,6–10 The large size of the orbital exenteration defect, dural exposure, exposure of paranasal sinuses in addition to the orbital exenteration cavity, and the need for adjuvant postoperative high-dose radiation therapy dictate the use of large vascularized free flaps in some cases.6,7 If an orbital prosthesis is desired, a skin graft or fasciocutaneous flap may be placed, leaving an “open” cavity with a concave orbital socket.10 Placement of bulkier myocutaneous flaps results in a “closed” cavity.10 Because tumor recurrence following orbital exenteration can occur in 24% to 45% of cases4,5 and may not be evident on clinical examination, imaging surveillance is imperative for early detection and appropriate treatment of recurrence. To the best of the authors’ knowledge, the imaging characteristics of tumor recurrence following orbital exenteration and free flap reconstruction have not been previously reported. The purpose of this study was to describe the imaging characteristics of recurrent tumors of orbital area on CT, MRI, and positron emission tomography (PET/CT) scans in patients who had an orbital exenteration and a free flap reconstruction. The authors aimed to compare the imaging appearance of recurrent tumors with the normal appearance of reconstruction flaps described in the literature. They hypothesized that recurrent tumors have an appearance different from that of normal flaps.
MATERIALS AND METHODS The Institutional Review Board at The University of Texas MD Anderson Cancer Center approved this study and waived the requirement for informed consent. The authors reviewed the clinical data and imaging studies for 48 consecutive patients who had orbital exenteration and free flap reconstruction at their institution between May 2005 and July 2011. From this group of 48 patients, the authors selected for inclusion in the current study patients who had recurrent tumors diagnosed in or adjacent to the operative site after an orbital exenteration. CT scanning was performed on GE scanners (Milwaukee, WI, U.S.A.) after administration of intravenous contrast material (Optiray, Mallinckrodt Inc., St. Louis, MO, U.S.A.). The following parameters were used for the CT studies: slice thickness 1.25 to 5 mm, field of
315
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
view 180 to 250, kVp 120 to 140, mA 180 to 220. MR images were obtained using a GE 1.5-T unit and included pre- and postgadolinium T1-weighted and T2-weighted images. Intravenous gadolinium (Omniscan, GE Healthcare) was administered in all cases. Imaging was performed at a slice thickness of 3 to 6 mm with a slice gap of 1 to 2 mm. Fluorodeoxyglucose (FDG) PET/CT scans were performed on a dedicated PET/CT system (Discovery ST, STe, or RX; GE Medical Systems). Scans were acquired from the orbits through the midthighs. Scans were acquired 60 to 90 minutes after intravenous administration of 18FDG at a dose of 8.1 to 19 mCi (mean, 15.9 mCi). PET studies were acquired in either 2-dimensional or 3-dimensional acquisition mode at 3 to 5 minutes per bed position (depending on the patient’s body mass index). The CT, MRI, and PET studies were reviewed by head and neck radiologists (J.M.D. and P.S.L.) with 12-year and 1-year experience, respectively; the lesions were assessed for the presence of a soft tissue mass, imaging characteristics of the mass, bone destruction, and FDG avidity.
RESULTS In the group of 48 patients who underwent orbital exenteration with free flap reconstruction, 17 recurrent tumors were diagnosed in 17 patients (35%; 11 men and 6 women) during the follow-up period. The patients with recurrent tumors ranged in age from 19 to 72 years (median, 59 years). Primary Tumor Characteristics and Treatments. All 17 patients underwent orbital exenteration for oncologic purposes (Table 1). The most common primary malignancies were squamous cell carcinoma and adenoid cystic carcinoma. All 17 patients also underwent flap-based reconstruction following resection of the primary tumor, 2 of which were performed at an outside institution. The most common types of flaps used were anterolateral thigh and radial forearm flaps. Five patients were treated with adjuvant radiation (median, 60 Gy; range, 54–64 Gy). Frequency and Type of Imaging Studies Recurrent Tumors. The median time between exenteration and the detection of recurrent disease was 15 months (range, 1–82 months). The median time between completion of radiation therapy and the detection of recurrence was 22 months (range, 5–77 months). Ten patients experienced recurrence during the first 2 years after orbital exenteration. Two of these patients had recurrence on the initial imaging study, and the other 8 patients were followed with CT or MR between 2 and 4 months until the recur-
rence was detected. For the 7 patients in whom the recurrence was detected after 2 years, the available studies during the first 2 years after surgery ranged from every 4 months to once per year. After 2 years, the patients were followed every 6 months until development of the recurrence. Other imaging studies may be available but not submitted for the authors’ review. The recurrent tumors were located at the margin of the flap in 13 of the 17 patients (Table 2). In the remaining 4 patients, the recurrent tumor was remote from the operative site; the locations were the sinonasal cavity (n = 1), inferior orbital rim (n = 1), masticator space (n = 1), and infratemporal fossa (n = 1). In 10 patients, biopsies confirmed recurrent disease with pathologically similar findings to the primary tumor; however, the remaining 7 patients were treated for recurrence on the basis of the imaging findings alone. Ten of the 17 patients had symptoms at the time of presentation with recurrence. Primary signs and symptoms included pain at the recurrence site (n = 6), mass (n = 3), and swelling (n = 1). Six recurrent tumors were imaged with CT, and 11 were imaged with MRI (Table 3); patient 7 was also imaged with PET/CT. On imaging, all 17 recurrences presented as a soft tissue mass. The lesions ranged in size from 1.0 to 7.4 cm (median, 2.3 cm). In the 6 lesions imaged with CT, enhancement patterns of the soft tissue mass were as follows: mild (n = 1), homogeneous (n = 2), heterogeneous (n = 1), and centrally necrotic with mild peripheral nodular enhancement (n = 2); bone destruction was not present in any cases. The 11 lesions imaged with MRI had the following enhancement patterns: mild (n = 1), homogeneous (n = 5), heterogeneous (n = 3), and peripheral/centrally necrotic (n = 2). Lesions were T1 hypointense (n = 4) or isointense (n = 7) to muscle before contrast material was administered. On T2, 8 lesions were hypointense, 1 was isointense, and 2 were hyperintense. In the patient who underwent PET/CT, the standard uptake value of the recurrent tumor was 13.8 (Fig. 1). Baseline imaging of the primary tumor was available in 15 of 17 cases. The recurrent tumor had a similar imaging appearance in 12 of the 15 cases (Fig. 2). In the other 3 cases, 1 had similar enhancement, but more T2 signal intensity, 1 was larger and more necrotic, and the final lesion had perineural tumor spread (Fig. 3). Following the diagnosis of a recurrent tumor, patients were treated with chemotherapy alone (n = 6), surgical resection (n = 5), radiation alone (n = 2), or chemoradiation (n = 2) or were lost to follow up before treatment was initiated (n = 2). Overall, mean survival was 12 months (range, 4–42 months) following exenteration, with 10 patients succumbing to disease at the time of the study.
TABLE 1. Patient, primary tumor, and treatment details Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Age, years/sex
Primary tumor type
Primary tumor location
Type of orbital surgery/type of flap
45/M 50/M 69/F 31/M 25/F 62/M 71/F 72/F 19/M 72/F 50/M 58/F 69/M 66/M 65/M 59/M 46/M
Squamous cell carcinoma Adenocarcinoma Poorly differentiated carcinoma Adenoid cystic carcinoma Adenoid cystic carcinoma Squamous cell carcinoma Myofibroblastic sarcoma Basosquamous cell carcinoma Myoepithelial carcinoma Squamous cell carcinoma Adenoid cystic carcinoma Adenoid cystic carcinoma Squamous cell carcinoma Squamous cell carcinoma Malignant fibrous histiocytoma Adenocarcinoma Basal cell carcinoma
Left sinonasal cavity Right lacrimal sac/nasolacrimal duct Right lacrimal sac Right lacrimal gland Right lacrimal gland Right maxilla Left maxillary sinus Right medial canthus/lacrimal fossa Right orbit Left maxillary sinus Right lacrimal gland Right maxilla/orbital floor Left maxillary sinus Right orbit/ophthalmic nerve Left maxillary sinus, ethmoid sinus Right orbit Periorbital/sinonasal cavity
Exenteration + maxillectomy/VRAM flap Exenteration + maxillectomy/radial forearm flap Exenteration + maxillectomy/ALT flap Exenteration/radial forearm flap Exenteration/radial forearm flap Exenteration + maxillectomy/ALT flap Exenteration/fibula osteocutaneous flap Exenteration + maxillectomy/ALT flap Exenteration/ALT flap Exenteration + maxillectomy/ALT flap Exenteration/ radial forearm flap Exenteration/DIEP flap Exenteration/ALT flap Exenteration/ALT flap Maxillectomy + exenteration/ALT flap Exenteration/radial forearm flap Exenteration/ALT flap
ALT, anterolateral thigh; DIEP, deep inferior epigastric perforators; VRAM, vertical rectus abdominis myocutaneous.
316
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
Recurrence Following Orbital Exenteration
TABLE 2. Size, location, imaging appearance, and treatment of recurrent tumors and patient outcomes after treatment Patient
Size of recurrent tumor, cm
Location of recurrent tumor*
1 2 3 4
2.2 1.2 4.3 1.9
Posterior margin Inferior margin Superior margin Anterosuperior margin
5 6 7 8 9 10 11 12 13 14 15 16 17
2 6 1 1.1 5 7.4 2.2 2.7 5.5 1.5 4.3 1 2.5
Masseter muscle Inferolateral margin Posterior margin Medial margin Posterior margin† Inferior margin Posterior margin† Right infratemporal fossa Posterior margin† Superior margin Inferomedial margin Inferior orbital rim Sinonasal cavity
Appearance of recurrence similar to that of primary tumor?
Treatment for recurrence
Patient status at time of report
Yes Yes Yes No—homogeneous enhancement/T2 more intense NA NA Yes Yes Yes Yes Yes Yes Yes No—perineural spread No—larger/more necrotic Yes Yes
Chemotherapy Chemotherapy Chemotherapy Surgery
Dead Dead Dead Continue to follow
Surgery Chemotherapy Chemotherapy Surgery Radiation therapy Chemoradiation Radiation therapy Lost to follow up Lost to follow up Chemoradiation Chemotherapy Surgery Surgery
Dead Dead Dead Continue to follow Dead Dead Dead Dead Lost to follow up Continue to follow Continue to follow Continue to follow Continue to follow
*Margin indicates flap margin. †Included perineural spread. NA, not applicable.
TABLE 3. Findings on imaging of recurrent tumors Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Type of imaging
Imaging appearance
MRI CT CT MRI MRI MRI CT CT MRI CT MRI MRI MRI MRI CT MRI MRI
T1 isointense, T2 hypointense, heterogeneous enhancement Homogeneous enhancement Peripheral enhancement, centrally necrotic, mild enhancement T1 isointense, T2 heterogeneously hyperintense, heterogeneous enhancement T1 isointense, T2 hypointense, peripheral enhancement, centrally necrotic T1 isointense, T2 hypointense, peripheral enhancement, centrally necrotic Homogeneous enhancement Mild enhancement T1 isointense, T2 heterogeneously hypointense, homogeneous enhancement Peripheral enhancement, centrally necrotic T1 and T2 hypointense, homogeneous enhancement T1 hypointense, T2 hyperintense, homogeneous enhancement T1 and T2 hypointense, mild enhancement T1 and T2 hypointense, homogeneous enhancement Heterogeneous enhancement T1 isointense, T2 hypointense, heterogeneous enhancement T1 isointense, T2 isointense, homogeneous enhancement
DISCUSSION The findings demonstrate that tumors that recur following orbital exenteration have variable imaging appearances. The enhancement on CT and MRI ranged from mild to homogeneous to centrally necrotic with mild peripheral nodular enhancement. The recurrences were usually isointense to muscle on T1-weighted precontrast images and generally T2 hypointense. There was no specific pattern of enhancement or signal characteristic correlating with the histology of the recurrence. However, these recurrent tumors all appeared as a soft tissue mass, often similar in appearance to the primary tumor. The time from orbital exenteration to tumor recurrence also varied greatly, ranging from 1 month to almost 7 years, but the majority (provide 65%) of recurrences occurred within the first 2 years after orbital exenteration and free flap reconstruction. Most recurrences following flap reconstruction also occurred at the margin of the flap (13 of 17 cases), which is
similar to patterns of recurrence reported in previous clinical studies of primary11–16 and radiation-associated tumors.17 Comparison of the authors’ findings with previously reported information about the imaging appearance of flaps without recurrent disease used for orbital reconstruction provides some useful information. Sedrak et al. (presented at the American Society of Neuroradiology Annual Meeting, San Diego, 2013) described the appearance of normal flaps following orbital exenteration usually as T1 isointense to muscle and T2 hyperintense to muscle. Like normal flaps, most of the recurrent tumors in this study were T1 isointense to muscle. However, unlike normal flaps, the recurrent tumors in this study were usually T2 isointense and less often hyperintense. While the flaps in the study by Sedrak et al. demonstrated a variable degree of persistent enhancement, none of the flaps appeared as a soft tissue mass. In contrast, all the recurrent tumors in this study appeared as soft tissue masses.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
317
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
FIG. 1. A 71-year-old woman (patient 7) with myofibroblastic sarcoma of the left maxillary sinus. A, Axial contrastenhanced CT, soft tissue window; orbital exenteration with fibula osteocutaneous flap reconstruction, 2 months before recurrence (arrow). B, Axial contrast-enhanced CT, soft tissue window; subtle but homogeneous enhancing recurrence along posterior margin of the flap reconstruction (arrows). C, Coronal contrast-enhanced CT, soft tissue window. Subtle enhancement of the recurrence (arrow). D, Axial positron emission tomography/CT. FDG-avid recurrence along the posterior margin of the flap (SUV = 13.8; arrow). Chong et al.18 also reported the enhancing appearance of flaps used for reconstruction after the treatment of head and neck tumors as persistently enhancing but with variable T2 signal. An important feature of their study was the striated appearance of the native muscle in the myocutaneous flaps. For both fasciocutaneous and myocutaneous flaps, the presence of a new soft tissue mass and loss of the striated appearance of muscle should alert the interpreting physician to the possibility of recurrent tumor. Early detection of tumor recurrence in patients who have undergone orbital exenteration is an important task but is often difficult because of postoperative changes due to loss of normal anatomic landmarks and the bulkiness of the free flaps that are needed to cover large cavities after orbital exenteration particularly if the paranasal sinuses or brain tissue is exposed during the operation. Detection is made more challenging by the extensive reconstruction common in these patients. PET has been used to help achieve the commonly difficult task of differentiating recurrence from scarring or radiation-associated tissue changes.
318
Lane and Bilyk12 used PET in cases in which conventional imaging could not differentiate between postexenteration scarring and tumor recurrence. They reported that when the findings on PET were negative the patients did not have evidence of recurrence on subsequent conventional imaging or clinical examinations. Rho et al.19 used PET standardized uptake value (SUV) ratios (i.e., lesion to nonlesion, lesion to aorta, and lesion to cerebellum) to detect recurrence in patients with maxillary sinus cancer treated with maxillectomy and adjuvant radiation treatment, citing the difficulty of detecting recurrence after surgery and chemoradiation because of anatomic distortion and posttreatment changes. The case in this study for which PET/ CT was performed had increased FDG uptake (SUV 13.8) and was easily detectable (Fig. 1). Further study will be needed to determine the role of PET/CT in this setting. Guerra et al.6 described the advantages and disadvantages of a temporalis flap versus spontaneous granulation and skin grafting following exenteration, providing some insight in the difficulty of clinical diagnosis of recurrence. They stated that
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
Recurrence Following Orbital Exenteration
FIG. 2. A 46-year-old man (patient 17) with basal cell carcinoma of the right sinonasal cavity and orbit, status post surgical resection, orbital exenteration, and anterolateral thigh flap reconstruction. A, Coronal T1 postcontrast. Homogeneously enhancing mass is present in the right sinonasal cavity with extension in the orbit (arrows). B and C, Axial T1 precontrast and axial FSE T2. Isointense recurrent tumor in the anterior ethmoid air cells (arrow). D, Axial T1 postcontrast. Recurrent tumor with a similar homogeneous enhancing appearance to the primary tumor (arrow). while flap-based reconstruction is associated with better resistance to infection and better tolerance of radiation treatment, flaps are more difficult to survey for recurrent tumor. In this series, all 17 patients had reconstruction with a flap, and 7 of 17 patients (41%) were asymptomatic at the time the recurrent tumor was detected by imaging. This illustrates the importance of accurate imaging surveillance in detecting tumor recurrence. Reported recurrence and survival rates for patients who have undergone exenteration of orbital tumors are variable. Bartley et al.20 reported on 19 different types of neoplasms in 102 patients. Squamous cell carcinoma, basal cell carcinoma, and melanoma accounted for 70% of the neoplasms. In 82 patients with no known residual tumor or metastases after surgery, the 1-year survival rate was 88.6%, and the 5-year survival rate was 56.8%. The 5-year rate of survival free of recurrence or metastases was 48.3%. Savage21 in a study that included 11 patients who underwent orbital exenteration with reconstruction including musculocutaneous flaps and free tissue transfers for extensive squamous or basal cell carcinomas of the skin reported that the recurrence rate was 60%, and the 5-year
survival rate was 56%. Kuo et al.4 reported that 35 (92%) of the 38 patients underwent exenteration with flap reconstructions for treatment of orbital tumors, including squamous cell carcinoma (n = 19), basal cell carcinoma (n = 12), and melanoma (n = 5), melanocytoma (n = 1), and chondrosarcoma (n = 1). In their study, the disease-specific survival rate was 97% at 1 year and 92% at 5 years. The local control rate was 83% at 1 year and 55% at 5 years. While it is outside the scope of the current article to discuss recurrence rates in this cohort, the recurrence rate was 35% (17 of 48 patients), and 10 of 17 patients (59%) with recurrent disease were alive at 1 year following exenteration. Differences between these studies and this study may be due to the small sample size and the more aggressive and multiply recurrent tumor types that are typical in the patient population.18,22–24 Seven of the 48 patients (15%) originally reviewed for possible recurrence following orbital exenteration had adenoid cystic carcinoma, 6 involved the lacrimal gland, and 1 the maxillary sinus. Following exenteration, 4 of the 7 patients (57%) patients had recurrence, 3 of which initially involved the lacrimal gland. In 1 of these 3 patients, there was perineural
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
319
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
FIG. 3. A 50-year-old man (patient 11) with adenoid cystic carcinoma of the right lacrimal gland, status post radial forearm flap reconstruction. A, Axial T1 postcontrast. Primary tumor in the right lacrimal gland and orbit (arrows). B, Axial T1 postcontrast, 13 months after surgery. Homogeneously enhancing mass is present along the posterior margin of the flap with perineural spread to involve the cavernous sinus (arrows). C, Coronal T1 postcontrast. Enlargement of the right cavernous sinus consistent with perineural spread (arrow). spread at presentation, and 1 patient later developed distant metastasis. Of the 10 deaths among patients with recurrence, 3 were from adenoid cystic carcinoma. At the authors’ institution, “baseline” imaging is obtained at 6 weeks to 3 months after exenteration and free flap reconstruction. After the first “baseline” study, the authors image every 3 to 4 months for the first year, every 6 months the second year, and at longer intervals for years 3 to 5 after exenteration. The type of imaging, either CT or MRI, is left to the discretion of the ocular surgeon and the interpreting radiologist.
CONCLUSIONS Recurrences after orbital exenteration that may occur months to several years after treatment are often clinically aggressive and challenging to detect on physical examination. These recurrences are often detected initially on imaging and appear as a mass, with imaging characteristics similar to those of the primary tumor. On both CT and MRI, the enhancing patterns range from mild to homogeneous to centrally necrotic with mild peripheral nodular enhancement. However, on T2-weighted images, the recurrent tumors are often T2 isointense, as opposed to the hyperintense appearance described in normal flaps. The appearance of the recurrent tumor is different from that of the native flap, and recurrences often occur at the margin of the flap. Head and neck radiologists, orbital and plastic surgeons, and orbital oncologists who care for patients following orbital exenteration should be aware of the wide range of clinical and imaging presentations of postexenteration tumor recurrence because failure to detect the recurrence early may lead to delay of appropriate treatment. The authors recommend regular surveillance using imaging studies to entail every 3 to 4 months of imaging during the first 2 years of follow up and less frequently afterward.
ACKNOWLEDGMENT
The authors thank Stephanie Deming for assistance in the preparation of this manuscript.
REFERENCES 1. Weisman RA, Kikkawa D, Moe KS, et al. Orbital tumors. Otolaryngol Clin North Am 2001;34:1157–74. 2. Tyers AG. Orbital exenteration for invasive skin tumors. Eye 2006;20:1165–70.
320
3. Suárez C, Ferlito A, Lund VJ, et al. Management of the orbit in malignant sinonasal tumors. Head Neck 2008;30:242–50. 4. Kuo CH, Gao K, Clifford A, et al. Orbital exenterations: an 18-year experience from a single head and neck unit. ANZ J Surg 2011;81:326–30. 5. Nemet AY, Martin P, Benger R, et al. Orbital exenteration: a 15-year study of 38 cases. Ophthal Plast Reconstr Surg 2007;23:468–72. 6. Guerra AS, Barbosa R, Choupina M, et al. Orbital exenteration for eyelid skin carcinoma. Eur J Plast Surg 2011:34:239–43. 7. Croce A, Moretti A, D’Agostino L, et al. Orbital exenteration in elderly patients: personal experience. Acta Otorhinolaryngol Ital 2008;28:193–9. 8. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy: part I. Algorithm and outcomes. Plast Reconstr Surg 2012;129:124–36. 9. Weichel ED, Eiseman AS, Casler JD, et al. Rectus abdominis free flap in the reconstruction of the orbit following subtotal exenteration. Ophthalmic Surg Lasers Imaging 2011;42:83–6. 10. Hanasono MM, Lee JC, Yang JS, et al. An algorithmic approach to reconstructive surgery and prosthetic rehabilitation after orbital exenteration. Plast Reconstr Surg 2009;123:98–105. 11. Laskar S, Basu A, Muckaden MA, et al. Osteosarcoma of the head and neck region: lessons learned from a single-institution experience of 50 patients. Head Neck 2008;30:1020–6. 12. Lane KA, Bilyk JR. Preliminary study of positron emission tomography in the detection and management of orbital malignancy. Ophthal Plast Reconstr Surg 2006;22:361–5. 13. Hudgins PA, Burson JG, Gussack GS, et al. CT and MR appearance of recurrent malignant head and neck neoplasms after resection and flap reconstruction. AJNR Am J Neuroradiol 1994;15:1689–94. 14. Hudgins PA. Flap reconstruction in the head and neck: expected appearance, complications, and recurrent disease. Eur J Radiol 2002;44:130–8. 15. Tomura N, Watanabe O, Hirano Y, et al. MR imaging of recurrent tumours following reconstructive surgery. Clin Radiol 2002;57:109–13. 16. Amin AA, Sakkary MA, Khalil AA, et al. The submental flap for oral cavity reconstruction: extended indications and technical refinements. Head Neck Oncol 2011;3:51. 17. Debnam JM, Guha-Thakurta N, Mahfouz YM, et al. Radiation-associated head and neck sarcomas: spectrum of imaging findings. Oral Oncol 2012;48:155–61. 18. Chong J, Chan LL, Langstein HN, et al. MR imaging of the muscular component of myocutaneous flaps in the head and neck. AJNR Am J Neuroradiol 2001;22:170–4. 19. Rho HJ, Kim SJ, Nam HY, et al. Detection and prediction of local recurrence of maxillary sinus cancer using F-18 FDG PET/CT. Eur J Surg Oncol 2010;36:214–20.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
20. Bartley GB, Garrity JA, Waller RR, et al. Orbital exenteration at the Mayo Clinic. 1967-1986. Ophthalmology 1989;96:468–73. 21. Savage RC. Orbital exenteration and reconstruction for mas sive basal cell and squamous cell carcinoma of cutaneous origin. Ann Plast Surg 1983;10:458–66. 22. Esmaeli B, Ahmadi MA, Youssef A, et al. Outcomes in patients with adenoid cystic carcinoma of the lacrimal gland. Ophthal Plast Reconstr Surg 2004;20:22–6.
Recurrence Following Orbital Exenteration
23. Ahmad SM, Esmaeli B, Williams M, et al. American Joint Committee on Cancer classification predicts outcome of patients with lacrimal gland adenoid cystic carcinoma. Ophthalmology 2009;116:1210–5. 24. Williams MD, Al-Zubidi N, Debnam JM, et al. Bone invasion by adenoid cystic carcinoma of the lacrimal gland: preoperative imaging assessment and surgical considerations. Ophthal Plast Reconstr Surg 2010;26:403–8.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
321
Imaging Findings of Recurrent Tumors After Orbital Exenteration and Free Flap Reconstruction Paul S. Lee, M.D.*, Peter Sedrak, M.D.*, Nandita Guha-Thakurta, M.D.†, Edward I. Chang, M.D.‡, Lawrence E. Ginsberg, M.D.†, Bita Esmaeli, M.D.‡§, and James Matthew Debnam, M.D.† *Section of Neuroradiology, Department of Diagnostic Radiology, The University of Texas at Houston, Houston, Texas, U.S.A.; and †Section of Neuroradiology, Department of Diagnostic Radiology, ‡Department of Plastic Surgery, and §Orbital Oncology and Oculoplastic Surgery Program, The University of Texas MD Anderson Cancer Center, Houston, Texas, U.S.A.
Purpose: Tumors that recur following orbital exenteration may not be evident on clinical examination, highlighting the need for imaging surveillance. The goal of this study was to report the imaging characteristics of recurrent tumors following orbital exenteration and free flap reconstruction. Methods: The authors retrospectively reviewed the records of 48 patients who underwent orbital exenteration for the treatment of orbital malignancy and identified 17 recurrent tumors in 17 patients. The lesions were assessed for the presence of a soft tissue mass, imaging characteristics, and fluorodeoxyglucose avidity. Results: The recurrent tumors were detected 1 month to 6 years 10 months (median, 1 year 3 month) after orbital exenteration. On both CT and MRI, all 17 lesions were soft tissue masses at presentation. On CT, the lesions demonstrated heterogeneous to homogeneous to centrally necrotic enhancement; on MRI, the lesions were T1 hypointense to isointense and T2 hypointense to hyperintense. Twelve of the 15 recurrent tumors with available preoperative imaging had an enhancing appearance similar to that of the original tumor. Thirteen of the 17 recurrent tumors were at the margin of a flap placed for reconstruction; the other 4 lesions were remote from the operative site. Conclusion: Recurrent tumors following orbital exenteration and free flap reconstruction demonstrate a wide range of imaging appearances but most often appear as a soft tissue masses often similar in appearance to the primary tumor and arising near the flap margin. Awareness of the imaging features of recurrent disease is important because failure to diagnose recurrence can delay appropriate treatment. (Ophthal Plast Reconstr Surg. 2014;30:315–321)
O
rbital tumors arise from the structures within the orbit and periorbital region. Primary orbital tumors include lymphoproliferative tumors, mesenchymal tumors, and tumors of the lacrimal gland, including adenoid cystic carcinoma. Tumors Accepted for publication November 23, 2013. Supported by The University of Texas MD Anderson Cancer Center is supported, in part, by the National Institutes of Health through Cancer Center Support Grant CA016672. The authors have no conflicts of interest to disclose. Address correspondence and reprint requests to James Matthew Debnam, M.D., The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Unit 1482, Houston, TX 77030. E-mail: matthew.debnam@ mdanderson.org DOI: 10.1097/IOP.0000000000000100
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
arising from the paranasal sinuses can secondarily involve the orbit.1 Other primary tumors can also metastasize to the orbit. Treatment of aggressive orbital tumors may include orbital exenteration, a radical procedure that involves the removal of the soft tissue contents of the orbit, including the globe, extraocular muscle and optic nerve, eyelids, and adjacent skin. Orbital exenteration is usually reserved for aggressive orbital tumors or locally advanced sinonasal malignancies that are potentially fatal and relentlessly progressive.2–6 A variety of types of flaps and grafts can be used for reconstruction following exenteration.4,6–10 The large size of the orbital exenteration defect, dural exposure, exposure of paranasal sinuses in addition to the orbital exenteration cavity, and the need for adjuvant postoperative high-dose radiation therapy dictate the use of large vascularized free flaps in some cases.6,7 If an orbital prosthesis is desired, a skin graft or fasciocutaneous flap may be placed, leaving an “open” cavity with a concave orbital socket.10 Placement of bulkier myocutaneous flaps results in a “closed” cavity.10 Because tumor recurrence following orbital exenteration can occur in 24% to 45% of cases4,5 and may not be evident on clinical examination, imaging surveillance is imperative for early detection and appropriate treatment of recurrence. To the best of the authors’ knowledge, the imaging characteristics of tumor recurrence following orbital exenteration and free flap reconstruction have not been previously reported. The purpose of this study was to describe the imaging characteristics of recurrent tumors of orbital area on CT, MRI, and positron emission tomography (PET/CT) scans in patients who had an orbital exenteration and a free flap reconstruction. The authors aimed to compare the imaging appearance of recurrent tumors with the normal appearance of reconstruction flaps described in the literature. They hypothesized that recurrent tumors have an appearance different from that of normal flaps.
MATERIALS AND METHODS The Institutional Review Board at The University of Texas MD Anderson Cancer Center approved this study and waived the requirement for informed consent. The authors reviewed the clinical data and imaging studies for 48 consecutive patients who had orbital exenteration and free flap reconstruction at their institution between May 2005 and July 2011. From this group of 48 patients, the authors selected for inclusion in the current study patients who had recurrent tumors diagnosed in or adjacent to the operative site after an orbital exenteration. CT scanning was performed on GE scanners (Milwaukee, WI, U.S.A.) after administration of intravenous contrast material (Optiray, Mallinckrodt Inc., St. Louis, MO, U.S.A.). The following parameters were used for the CT studies: slice thickness 1.25 to 5 mm, field of
315
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
view 180 to 250, kVp 120 to 140, mA 180 to 220. MR images were obtained using a GE 1.5-T unit and included pre- and postgadolinium T1-weighted and T2-weighted images. Intravenous gadolinium (Omniscan, GE Healthcare) was administered in all cases. Imaging was performed at a slice thickness of 3 to 6 mm with a slice gap of 1 to 2 mm. Fluorodeoxyglucose (FDG) PET/CT scans were performed on a dedicated PET/CT system (Discovery ST, STe, or RX; GE Medical Systems). Scans were acquired from the orbits through the midthighs. Scans were acquired 60 to 90 minutes after intravenous administration of 18FDG at a dose of 8.1 to 19 mCi (mean, 15.9 mCi). PET studies were acquired in either 2-dimensional or 3-dimensional acquisition mode at 3 to 5 minutes per bed position (depending on the patient’s body mass index). The CT, MRI, and PET studies were reviewed by head and neck radiologists (J.M.D. and P.S.L.) with 12-year and 1-year experience, respectively; the lesions were assessed for the presence of a soft tissue mass, imaging characteristics of the mass, bone destruction, and FDG avidity.
RESULTS In the group of 48 patients who underwent orbital exenteration with free flap reconstruction, 17 recurrent tumors were diagnosed in 17 patients (35%; 11 men and 6 women) during the follow-up period. The patients with recurrent tumors ranged in age from 19 to 72 years (median, 59 years). Primary Tumor Characteristics and Treatments. All 17 patients underwent orbital exenteration for oncologic purposes (Table 1). The most common primary malignancies were squamous cell carcinoma and adenoid cystic carcinoma. All 17 patients also underwent flap-based reconstruction following resection of the primary tumor, 2 of which were performed at an outside institution. The most common types of flaps used were anterolateral thigh and radial forearm flaps. Five patients were treated with adjuvant radiation (median, 60 Gy; range, 54–64 Gy). Frequency and Type of Imaging Studies Recurrent Tumors. The median time between exenteration and the detection of recurrent disease was 15 months (range, 1–82 months). The median time between completion of radiation therapy and the detection of recurrence was 22 months (range, 5–77 months). Ten patients experienced recurrence during the first 2 years after orbital exenteration. Two of these patients had recurrence on the initial imaging study, and the other 8 patients were followed with CT or MR between 2 and 4 months until the recur-
rence was detected. For the 7 patients in whom the recurrence was detected after 2 years, the available studies during the first 2 years after surgery ranged from every 4 months to once per year. After 2 years, the patients were followed every 6 months until development of the recurrence. Other imaging studies may be available but not submitted for the authors’ review. The recurrent tumors were located at the margin of the flap in 13 of the 17 patients (Table 2). In the remaining 4 patients, the recurrent tumor was remote from the operative site; the locations were the sinonasal cavity (n = 1), inferior orbital rim (n = 1), masticator space (n = 1), and infratemporal fossa (n = 1). In 10 patients, biopsies confirmed recurrent disease with pathologically similar findings to the primary tumor; however, the remaining 7 patients were treated for recurrence on the basis of the imaging findings alone. Ten of the 17 patients had symptoms at the time of presentation with recurrence. Primary signs and symptoms included pain at the recurrence site (n = 6), mass (n = 3), and swelling (n = 1). Six recurrent tumors were imaged with CT, and 11 were imaged with MRI (Table 3); patient 7 was also imaged with PET/CT. On imaging, all 17 recurrences presented as a soft tissue mass. The lesions ranged in size from 1.0 to 7.4 cm (median, 2.3 cm). In the 6 lesions imaged with CT, enhancement patterns of the soft tissue mass were as follows: mild (n = 1), homogeneous (n = 2), heterogeneous (n = 1), and centrally necrotic with mild peripheral nodular enhancement (n = 2); bone destruction was not present in any cases. The 11 lesions imaged with MRI had the following enhancement patterns: mild (n = 1), homogeneous (n = 5), heterogeneous (n = 3), and peripheral/centrally necrotic (n = 2). Lesions were T1 hypointense (n = 4) or isointense (n = 7) to muscle before contrast material was administered. On T2, 8 lesions were hypointense, 1 was isointense, and 2 were hyperintense. In the patient who underwent PET/CT, the standard uptake value of the recurrent tumor was 13.8 (Fig. 1). Baseline imaging of the primary tumor was available in 15 of 17 cases. The recurrent tumor had a similar imaging appearance in 12 of the 15 cases (Fig. 2). In the other 3 cases, 1 had similar enhancement, but more T2 signal intensity, 1 was larger and more necrotic, and the final lesion had perineural tumor spread (Fig. 3). Following the diagnosis of a recurrent tumor, patients were treated with chemotherapy alone (n = 6), surgical resection (n = 5), radiation alone (n = 2), or chemoradiation (n = 2) or were lost to follow up before treatment was initiated (n = 2). Overall, mean survival was 12 months (range, 4–42 months) following exenteration, with 10 patients succumbing to disease at the time of the study.
TABLE 1. Patient, primary tumor, and treatment details Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Age, years/sex
Primary tumor type
Primary tumor location
Type of orbital surgery/type of flap
45/M 50/M 69/F 31/M 25/F 62/M 71/F 72/F 19/M 72/F 50/M 58/F 69/M 66/M 65/M 59/M 46/M
Squamous cell carcinoma Adenocarcinoma Poorly differentiated carcinoma Adenoid cystic carcinoma Adenoid cystic carcinoma Squamous cell carcinoma Myofibroblastic sarcoma Basosquamous cell carcinoma Myoepithelial carcinoma Squamous cell carcinoma Adenoid cystic carcinoma Adenoid cystic carcinoma Squamous cell carcinoma Squamous cell carcinoma Malignant fibrous histiocytoma Adenocarcinoma Basal cell carcinoma
Left sinonasal cavity Right lacrimal sac/nasolacrimal duct Right lacrimal sac Right lacrimal gland Right lacrimal gland Right maxilla Left maxillary sinus Right medial canthus/lacrimal fossa Right orbit Left maxillary sinus Right lacrimal gland Right maxilla/orbital floor Left maxillary sinus Right orbit/ophthalmic nerve Left maxillary sinus, ethmoid sinus Right orbit Periorbital/sinonasal cavity
Exenteration + maxillectomy/VRAM flap Exenteration + maxillectomy/radial forearm flap Exenteration + maxillectomy/ALT flap Exenteration/radial forearm flap Exenteration/radial forearm flap Exenteration + maxillectomy/ALT flap Exenteration/fibula osteocutaneous flap Exenteration + maxillectomy/ALT flap Exenteration/ALT flap Exenteration + maxillectomy/ALT flap Exenteration/ radial forearm flap Exenteration/DIEP flap Exenteration/ALT flap Exenteration/ALT flap Maxillectomy + exenteration/ALT flap Exenteration/radial forearm flap Exenteration/ALT flap
ALT, anterolateral thigh; DIEP, deep inferior epigastric perforators; VRAM, vertical rectus abdominis myocutaneous.
316
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
Recurrence Following Orbital Exenteration
TABLE 2. Size, location, imaging appearance, and treatment of recurrent tumors and patient outcomes after treatment Patient
Size of recurrent tumor, cm
Location of recurrent tumor*
1 2 3 4
2.2 1.2 4.3 1.9
Posterior margin Inferior margin Superior margin Anterosuperior margin
5 6 7 8 9 10 11 12 13 14 15 16 17
2 6 1 1.1 5 7.4 2.2 2.7 5.5 1.5 4.3 1 2.5
Masseter muscle Inferolateral margin Posterior margin Medial margin Posterior margin† Inferior margin Posterior margin† Right infratemporal fossa Posterior margin† Superior margin Inferomedial margin Inferior orbital rim Sinonasal cavity
Appearance of recurrence similar to that of primary tumor?
Treatment for recurrence
Patient status at time of report
Yes Yes Yes No—homogeneous enhancement/T2 more intense NA NA Yes Yes Yes Yes Yes Yes Yes No—perineural spread No—larger/more necrotic Yes Yes
Chemotherapy Chemotherapy Chemotherapy Surgery
Dead Dead Dead Continue to follow
Surgery Chemotherapy Chemotherapy Surgery Radiation therapy Chemoradiation Radiation therapy Lost to follow up Lost to follow up Chemoradiation Chemotherapy Surgery Surgery
Dead Dead Dead Continue to follow Dead Dead Dead Dead Lost to follow up Continue to follow Continue to follow Continue to follow Continue to follow
*Margin indicates flap margin. †Included perineural spread. NA, not applicable.
TABLE 3. Findings on imaging of recurrent tumors Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Type of imaging
Imaging appearance
MRI CT CT MRI MRI MRI CT CT MRI CT MRI MRI MRI MRI CT MRI MRI
T1 isointense, T2 hypointense, heterogeneous enhancement Homogeneous enhancement Peripheral enhancement, centrally necrotic, mild enhancement T1 isointense, T2 heterogeneously hyperintense, heterogeneous enhancement T1 isointense, T2 hypointense, peripheral enhancement, centrally necrotic T1 isointense, T2 hypointense, peripheral enhancement, centrally necrotic Homogeneous enhancement Mild enhancement T1 isointense, T2 heterogeneously hypointense, homogeneous enhancement Peripheral enhancement, centrally necrotic T1 and T2 hypointense, homogeneous enhancement T1 hypointense, T2 hyperintense, homogeneous enhancement T1 and T2 hypointense, mild enhancement T1 and T2 hypointense, homogeneous enhancement Heterogeneous enhancement T1 isointense, T2 hypointense, heterogeneous enhancement T1 isointense, T2 isointense, homogeneous enhancement
DISCUSSION The findings demonstrate that tumors that recur following orbital exenteration have variable imaging appearances. The enhancement on CT and MRI ranged from mild to homogeneous to centrally necrotic with mild peripheral nodular enhancement. The recurrences were usually isointense to muscle on T1-weighted precontrast images and generally T2 hypointense. There was no specific pattern of enhancement or signal characteristic correlating with the histology of the recurrence. However, these recurrent tumors all appeared as a soft tissue mass, often similar in appearance to the primary tumor. The time from orbital exenteration to tumor recurrence also varied greatly, ranging from 1 month to almost 7 years, but the majority (provide 65%) of recurrences occurred within the first 2 years after orbital exenteration and free flap reconstruction. Most recurrences following flap reconstruction also occurred at the margin of the flap (13 of 17 cases), which is
similar to patterns of recurrence reported in previous clinical studies of primary11–16 and radiation-associated tumors.17 Comparison of the authors’ findings with previously reported information about the imaging appearance of flaps without recurrent disease used for orbital reconstruction provides some useful information. Sedrak et al. (presented at the American Society of Neuroradiology Annual Meeting, San Diego, 2013) described the appearance of normal flaps following orbital exenteration usually as T1 isointense to muscle and T2 hyperintense to muscle. Like normal flaps, most of the recurrent tumors in this study were T1 isointense to muscle. However, unlike normal flaps, the recurrent tumors in this study were usually T2 isointense and less often hyperintense. While the flaps in the study by Sedrak et al. demonstrated a variable degree of persistent enhancement, none of the flaps appeared as a soft tissue mass. In contrast, all the recurrent tumors in this study appeared as soft tissue masses.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
317
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
FIG. 1. A 71-year-old woman (patient 7) with myofibroblastic sarcoma of the left maxillary sinus. A, Axial contrastenhanced CT, soft tissue window; orbital exenteration with fibula osteocutaneous flap reconstruction, 2 months before recurrence (arrow). B, Axial contrast-enhanced CT, soft tissue window; subtle but homogeneous enhancing recurrence along posterior margin of the flap reconstruction (arrows). C, Coronal contrast-enhanced CT, soft tissue window. Subtle enhancement of the recurrence (arrow). D, Axial positron emission tomography/CT. FDG-avid recurrence along the posterior margin of the flap (SUV = 13.8; arrow). Chong et al.18 also reported the enhancing appearance of flaps used for reconstruction after the treatment of head and neck tumors as persistently enhancing but with variable T2 signal. An important feature of their study was the striated appearance of the native muscle in the myocutaneous flaps. For both fasciocutaneous and myocutaneous flaps, the presence of a new soft tissue mass and loss of the striated appearance of muscle should alert the interpreting physician to the possibility of recurrent tumor. Early detection of tumor recurrence in patients who have undergone orbital exenteration is an important task but is often difficult because of postoperative changes due to loss of normal anatomic landmarks and the bulkiness of the free flaps that are needed to cover large cavities after orbital exenteration particularly if the paranasal sinuses or brain tissue is exposed during the operation. Detection is made more challenging by the extensive reconstruction common in these patients. PET has been used to help achieve the commonly difficult task of differentiating recurrence from scarring or radiation-associated tissue changes.
318
Lane and Bilyk12 used PET in cases in which conventional imaging could not differentiate between postexenteration scarring and tumor recurrence. They reported that when the findings on PET were negative the patients did not have evidence of recurrence on subsequent conventional imaging or clinical examinations. Rho et al.19 used PET standardized uptake value (SUV) ratios (i.e., lesion to nonlesion, lesion to aorta, and lesion to cerebellum) to detect recurrence in patients with maxillary sinus cancer treated with maxillectomy and adjuvant radiation treatment, citing the difficulty of detecting recurrence after surgery and chemoradiation because of anatomic distortion and posttreatment changes. The case in this study for which PET/ CT was performed had increased FDG uptake (SUV 13.8) and was easily detectable (Fig. 1). Further study will be needed to determine the role of PET/CT in this setting. Guerra et al.6 described the advantages and disadvantages of a temporalis flap versus spontaneous granulation and skin grafting following exenteration, providing some insight in the difficulty of clinical diagnosis of recurrence. They stated that
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
Recurrence Following Orbital Exenteration
FIG. 2. A 46-year-old man (patient 17) with basal cell carcinoma of the right sinonasal cavity and orbit, status post surgical resection, orbital exenteration, and anterolateral thigh flap reconstruction. A, Coronal T1 postcontrast. Homogeneously enhancing mass is present in the right sinonasal cavity with extension in the orbit (arrows). B and C, Axial T1 precontrast and axial FSE T2. Isointense recurrent tumor in the anterior ethmoid air cells (arrow). D, Axial T1 postcontrast. Recurrent tumor with a similar homogeneous enhancing appearance to the primary tumor (arrow). while flap-based reconstruction is associated with better resistance to infection and better tolerance of radiation treatment, flaps are more difficult to survey for recurrent tumor. In this series, all 17 patients had reconstruction with a flap, and 7 of 17 patients (41%) were asymptomatic at the time the recurrent tumor was detected by imaging. This illustrates the importance of accurate imaging surveillance in detecting tumor recurrence. Reported recurrence and survival rates for patients who have undergone exenteration of orbital tumors are variable. Bartley et al.20 reported on 19 different types of neoplasms in 102 patients. Squamous cell carcinoma, basal cell carcinoma, and melanoma accounted for 70% of the neoplasms. In 82 patients with no known residual tumor or metastases after surgery, the 1-year survival rate was 88.6%, and the 5-year survival rate was 56.8%. The 5-year rate of survival free of recurrence or metastases was 48.3%. Savage21 in a study that included 11 patients who underwent orbital exenteration with reconstruction including musculocutaneous flaps and free tissue transfers for extensive squamous or basal cell carcinomas of the skin reported that the recurrence rate was 60%, and the 5-year
survival rate was 56%. Kuo et al.4 reported that 35 (92%) of the 38 patients underwent exenteration with flap reconstructions for treatment of orbital tumors, including squamous cell carcinoma (n = 19), basal cell carcinoma (n = 12), and melanoma (n = 5), melanocytoma (n = 1), and chondrosarcoma (n = 1). In their study, the disease-specific survival rate was 97% at 1 year and 92% at 5 years. The local control rate was 83% at 1 year and 55% at 5 years. While it is outside the scope of the current article to discuss recurrence rates in this cohort, the recurrence rate was 35% (17 of 48 patients), and 10 of 17 patients (59%) with recurrent disease were alive at 1 year following exenteration. Differences between these studies and this study may be due to the small sample size and the more aggressive and multiply recurrent tumor types that are typical in the patient population.18,22–24 Seven of the 48 patients (15%) originally reviewed for possible recurrence following orbital exenteration had adenoid cystic carcinoma, 6 involved the lacrimal gland, and 1 the maxillary sinus. Following exenteration, 4 of the 7 patients (57%) patients had recurrence, 3 of which initially involved the lacrimal gland. In 1 of these 3 patients, there was perineural
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
319
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
P. S. Lee et al.
FIG. 3. A 50-year-old man (patient 11) with adenoid cystic carcinoma of the right lacrimal gland, status post radial forearm flap reconstruction. A, Axial T1 postcontrast. Primary tumor in the right lacrimal gland and orbit (arrows). B, Axial T1 postcontrast, 13 months after surgery. Homogeneously enhancing mass is present along the posterior margin of the flap with perineural spread to involve the cavernous sinus (arrows). C, Coronal T1 postcontrast. Enlargement of the right cavernous sinus consistent with perineural spread (arrow). spread at presentation, and 1 patient later developed distant metastasis. Of the 10 deaths among patients with recurrence, 3 were from adenoid cystic carcinoma. At the authors’ institution, “baseline” imaging is obtained at 6 weeks to 3 months after exenteration and free flap reconstruction. After the first “baseline” study, the authors image every 3 to 4 months for the first year, every 6 months the second year, and at longer intervals for years 3 to 5 after exenteration. The type of imaging, either CT or MRI, is left to the discretion of the ocular surgeon and the interpreting radiologist.
CONCLUSIONS Recurrences after orbital exenteration that may occur months to several years after treatment are often clinically aggressive and challenging to detect on physical examination. These recurrences are often detected initially on imaging and appear as a mass, with imaging characteristics similar to those of the primary tumor. On both CT and MRI, the enhancing patterns range from mild to homogeneous to centrally necrotic with mild peripheral nodular enhancement. However, on T2-weighted images, the recurrent tumors are often T2 isointense, as opposed to the hyperintense appearance described in normal flaps. The appearance of the recurrent tumor is different from that of the native flap, and recurrences often occur at the margin of the flap. Head and neck radiologists, orbital and plastic surgeons, and orbital oncologists who care for patients following orbital exenteration should be aware of the wide range of clinical and imaging presentations of postexenteration tumor recurrence because failure to detect the recurrence early may lead to delay of appropriate treatment. The authors recommend regular surveillance using imaging studies to entail every 3 to 4 months of imaging during the first 2 years of follow up and less frequently afterward.
ACKNOWLEDGMENT
The authors thank Stephanie Deming for assistance in the preparation of this manuscript.
REFERENCES 1. Weisman RA, Kikkawa D, Moe KS, et al. Orbital tumors. Otolaryngol Clin North Am 2001;34:1157–74. 2. Tyers AG. Orbital exenteration for invasive skin tumors. Eye 2006;20:1165–70.
320
3. Suárez C, Ferlito A, Lund VJ, et al. Management of the orbit in malignant sinonasal tumors. Head Neck 2008;30:242–50. 4. Kuo CH, Gao K, Clifford A, et al. Orbital exenterations: an 18-year experience from a single head and neck unit. ANZ J Surg 2011;81:326–30. 5. Nemet AY, Martin P, Benger R, et al. Orbital exenteration: a 15-year study of 38 cases. Ophthal Plast Reconstr Surg 2007;23:468–72. 6. Guerra AS, Barbosa R, Choupina M, et al. Orbital exenteration for eyelid skin carcinoma. Eur J Plast Surg 2011:34:239–43. 7. Croce A, Moretti A, D’Agostino L, et al. Orbital exenteration in elderly patients: personal experience. Acta Otorhinolaryngol Ital 2008;28:193–9. 8. Cordeiro PG, Chen CM. A 15-year review of midface reconstruction after total and subtotal maxillectomy: part I. Algorithm and outcomes. Plast Reconstr Surg 2012;129:124–36. 9. Weichel ED, Eiseman AS, Casler JD, et al. Rectus abdominis free flap in the reconstruction of the orbit following subtotal exenteration. Ophthalmic Surg Lasers Imaging 2011;42:83–6. 10. Hanasono MM, Lee JC, Yang JS, et al. An algorithmic approach to reconstructive surgery and prosthetic rehabilitation after orbital exenteration. Plast Reconstr Surg 2009;123:98–105. 11. Laskar S, Basu A, Muckaden MA, et al. Osteosarcoma of the head and neck region: lessons learned from a single-institution experience of 50 patients. Head Neck 2008;30:1020–6. 12. Lane KA, Bilyk JR. Preliminary study of positron emission tomography in the detection and management of orbital malignancy. Ophthal Plast Reconstr Surg 2006;22:361–5. 13. Hudgins PA, Burson JG, Gussack GS, et al. CT and MR appearance of recurrent malignant head and neck neoplasms after resection and flap reconstruction. AJNR Am J Neuroradiol 1994;15:1689–94. 14. Hudgins PA. Flap reconstruction in the head and neck: expected appearance, complications, and recurrent disease. Eur J Radiol 2002;44:130–8. 15. Tomura N, Watanabe O, Hirano Y, et al. MR imaging of recurrent tumours following reconstructive surgery. Clin Radiol 2002;57:109–13. 16. Amin AA, Sakkary MA, Khalil AA, et al. The submental flap for oral cavity reconstruction: extended indications and technical refinements. Head Neck Oncol 2011;3:51. 17. Debnam JM, Guha-Thakurta N, Mahfouz YM, et al. Radiation-associated head and neck sarcomas: spectrum of imaging findings. Oral Oncol 2012;48:155–61. 18. Chong J, Chan LL, Langstein HN, et al. MR imaging of the muscular component of myocutaneous flaps in the head and neck. AJNR Am J Neuroradiol 2001;22:170–4. 19. Rho HJ, Kim SJ, Nam HY, et al. Detection and prediction of local recurrence of maxillary sinus cancer using F-18 FDG PET/CT. Eur J Surg Oncol 2010;36:214–20.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Ophthal Plast Reconstr Surg, Vol. 30, No. 4, 2014
20. Bartley GB, Garrity JA, Waller RR, et al. Orbital exenteration at the Mayo Clinic. 1967-1986. Ophthalmology 1989;96:468–73. 21. Savage RC. Orbital exenteration and reconstruction for mas sive basal cell and squamous cell carcinoma of cutaneous origin. Ann Plast Surg 1983;10:458–66. 22. Esmaeli B, Ahmadi MA, Youssef A, et al. Outcomes in patients with adenoid cystic carcinoma of the lacrimal gland. Ophthal Plast Reconstr Surg 2004;20:22–6.
Recurrence Following Orbital Exenteration
23. Ahmad SM, Esmaeli B, Williams M, et al. American Joint Committee on Cancer classification predicts outcome of patients with lacrimal gland adenoid cystic carcinoma. Ophthalmology 2009;116:1210–5. 24. Williams MD, Al-Zubidi N, Debnam JM, et al. Bone invasion by adenoid cystic carcinoma of the lacrimal gland: preoperative imaging assessment and surgical considerations. Ophthal Plast Reconstr Surg 2010;26:403–8.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
321
