Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
Contents lists available at ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery journal homepage: www.jcmfs.com
Piezosurgery for orbital decompression surgery in thyroid associated orbitopathy Katharina A. Ponto a, e, *, Isabella Zwiener b, e, Bilal Al-Nawas c, George J. Kahaly d, Anna F. Otto a, Julia Karbach c, Norbert Pfeiffer a, Susanne Pitz a a
Dept. of Ophthalmology (Head: Prof. Dr. N. Pfeiffer), University Medical Center Mainz, Germany Dept. of Biostatistics, University Medical Center Mainz, Germany Dept. of Maxillofacial Surgery, University Medical Center Mainz, Germany d Medicine I, University Medical Center Mainz, Germany e Center of Thrombosis and Hemostasis (CTH), University Medical Center Mainz, Germany b c
a r t i c l e i n f o
a b s t r a c t
Article history: Paper received 9 January 2014 Accepted 18 June 2014 Available online 26 June 2014
The purpose of this study was to assess a piezosurgical device as a novel tool for bony orbital decompression surgery. At a multidisciplinary orbital center, 62 surgeries were performed in 40 patients with thyroid associated orbitopathy (TAO). Within this retrospective case-series, we analyzed the medical records of these consecutive unselected patients. The reduction of proptosis was the main outcome measure. Indications for a two (n ¼ 27, 44%) or three wall (35, 56%) decompression surgery were proptosis (n ¼ 50 orbits, 81%) and optic neuropathy (n ¼ 12, 19%). Piezosurgery enabled precise bone cuts without intraoperative complications. Proptosis decreased from 23.6 ± 2.8 mm (SD) by 3 mm (95% CI: 3.6 to 2.5 mm) after surgery and stayed stable at 3 months (3 mm, 95% CI: 3.61 to 2.5 mm, p < 0.001, respectively). The effect was higher in those with preoperatively higher values (>24 mm versus 24 mm: 3.4 mm versus 2.81 mm before discharge from hospital and 4.1 mm versus 2.1 mm at 3 months: p < 0.001, respectively). After a mean long-term follow-up period of 14.6 ± 10.4 months proptosis decreased by further 0.7 ± 2.0 mm (p < 0.001). Signs of optic nerve compression improved after surgery. Infraorbital hypesthesia was present in 11 of 21 (52%) orbits 3 months after surgery. The piezosurgical device is a useful tool for orbital decompression surgery in TAO. By cutting bone selectively, it is precise and reduces the invasiveness of surgery. Nevertheless, no improvement in outcome or reduction in morbidity over conventional techniques has been shown so far. © 2014 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
Keywords: Orbital surgery Decompression Thyroid associated orbitopathy Piezosurgery Proptosis
1. Introduction Thyroid associated orbitopathy (TAO) is the most frequent extrathyroidal manifestation of autoimmune hyperthyroidism (Bahn, 2010). In the inflammatory active phase of the disease, immunomodulatory treatment and/or retrobulbar irradiation are indicated (Chang and Piva, 2008; Douglas and Gupta, 2011; Kakizaki et al., 2011). In patients with severe but inactive TAO and/or in optic neuropathy orbital decompression surgery is
* Corresponding address. Dept. of Ophthalmology, University Medical Center Mainz, Langenbeckstr. 1, 55131 Mainz, Germany. Tel./fax: þ49 06131/17 7085x2290. E-mail addresses: [email protected], [email protected] (K.A. Ponto).
required (Bartalena et al., 2008; Baldeschi et al., 2010; Patel et al., 2012). In addition, decompression surgery may also be performed for prolonged orbital congestion, pain, or for cases with severe corneal exposure due to insufficient eyelid closure. As quality of life is markedly impaired in patients with severe proptosis (Robiony and Polini, 2010; Cassetta et al., 2012), cosmesis is increasingly recognized as an indication for orbital decompression. Nowadays, the swinging-eyelid approach has become the most popular method followed by the coronal and transconjunctival approaches (Paridaens et al., 2006; Baldeschi et al., 2010; Ponto et al., 2011, 2013; Zang et al., 2011). Nevertheless, in conventional surgery performed with hammer and chisel, damage to adjacent soft tissue is difficult to avoid, and the resulting bone fragments are barely predictable. Clinical studies especially from maxillofacial-, dental-, ENT- and neurosurgery have shown benefits of ultrasonic bone
http://dx.doi.org/10.1016/j.jcms.2014.06.020 1010-5182/© 2014 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
1814
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
cutting (Lytton et al., 2010; Ponto and Kahaly, 2010; Ponto et al., 2010; Zang et al., 2011). So far, studies on the piezosurgical device for orbital surgery are sparse and original data on efficacy and safety are lacking (Ponto et al., 2009). Due to the saw-like cutting properties modifications of the osteotomy procedure have to be established. We therefore aimed to evaluate this technique for bony orbital decompression surgery in patients with TAO at our institution. At the multidisciplinary orbital center of University Medical Center, Mainz, Germany, 62 bony orbital decompression surgeries with the piezosurgical device were performed in 40 patients with TAO between February 2010 and March 2012. Within this retrospective case-series, we analyzed he medical records or files of these consecutive unselected patients. The protocol was consistent with the principles of the Declaration of Helsinki. Because this retrospective case-series did not include any interventions aside from those commonly falling within the daily routine and because none of the individual-related data were passed to third parties, the responsible Ethics Committee of the Medical Association of the State Rhineland Palatinate, Germany, decided that no approval was required. 2. Material and methods 2.1. Surgical procedure Piezoelectric bone surgery (syn. Piezosurgery), is a technique of osteotomy that requires microvibrations of ultrasonic frequency scalpels. The principle of piezosurgery is ultrasonic transduction, obtained by piezoelectric ceramic contraction and expansion. The resulting vibrations are amplified and transferred onto the insert of a tip which results in the cavitation phenomenon with a
mechanical cutting effect, exclusively on mineralized tissues: with the piezosurgical device vibrating at 25e29 kHz, bones are cut selectively while sparing fine anatomical structures (e.g., vessels, nerve tissue), that would only be damaged when using higher frequencies. An oscillating tip with irrigation fluid enables effective cooling as well as higher visibility (via cavitation effect) (Schlee et al., 2006; Crosetti et al., 2009). The piezosurgical device we used (mectron™ piezosurgery, Cologne, Germany) consists of a basis unit, a surgical handpiece, and a footswitch. Irrigation and aspiration occur at the handpiece tip. For orbital decompression surgery, we used the saw-like tip (OT7) for ultrasonic bone cutting. The amplitude of tip vibration, rate of irrigation, and aspiration pressure can be regulated on the basis unit. A sterile saline solution is used for irrigation and cooling. A rate between 3 and 40 ml/min is used for irrigation. Before clinical use, the handling of the instrument for orbital decompression surgery was practiced on corpses of the institute of anatomy, Gutenberg University, Mainz, Germany (Fig. 1aed). The surgical procedure was personalized for each patient depending on the degree of proptosis, the presence of optic neuropathy, and the presence of eye motility disturbances. As a general principle, the decision about a two or three wall decompression surgery based on clinical severity: In the presence of optic neuropathy and high values of proptosis (cut-off: 22 mm) the effect of inferomedial decompression was implemented by adding lateral wall decompression (foremost of the inferior lateral orbital quadrant). We chose the relatively low cut-off of 22 mm because this value had been defined as a cut-off to dose the amount of decompression before (Paridaens et al., 2006) and because we aimed to perform a balanced decompression in most patients. Balanced orbital decompression (medial and lateral orbital wall with or without decompression of the orbital floor) has been
Fig. 1. Photographs at a practice session at the Institute of Anatomy. a: Exposition of the orbital floor via a transconjunctival approach with lateral canthotomy (“swinging eyelid”). b: Osteotomy with the piezosurgical device (OT7 tip). c: Bone removal laterally from the infraorbital nerve; an osteotomy medially from the nerve has already been performed. d: Removal of the medial segment of the orbital floor; the region of the infraorbital nerve is spared.
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
shown to be beneficial regarding the development of postoperative diplopia and motility disturbances (Goldberg et al., 2000; Unal et al., 2000; Kacker et al., 2003; Takahashi et al., 2008; Barkhuysen et al., 2009; Alsuhaibani et al., 2011). In patients with less proptosis, the lateral wall was not removed to avoid overcorrection (enophthalmos). All decompression surgeries were performed via a combined inferior fornix and transcaruncular approach. After conjunctival incision, the periosteum was incised along the inferior and lateral orbital rim. The periorbita was separated bluntly from the medial, inferior, and lateral orbital walls and incised in a posterioreanterior direction medially, infero-medially, and laterally. Approximately 1e1.5 cm of the anterior periorbita was left intact and herniation of fat was augmented through blunt dissection with scissors in the spaces between the rectus muscles. For decompression of the medial orbital wall we chose the transcaruncular approach as it is a valid alternative to the endoscopic approach (Shorr et al., 2000; Liao et al., 2006). It has been shown that decompression of the medial orbital wall successfully relieves the pressure at the orbital apex (McCann et al., 2006). The ethmoidal part of the medial orbital wall was decompressed as far posteriorly as reachable (approximately 1.5e2 cm) and as felt necessary for proptosis reduction. At our institution, we have gained positive experience with this method, even though it might be more difficult to quantify the amount of bone removal as it is possible under direct visualization using the endoscopic approach. Bone of the orbital floor was removed in two parts by cutting along the infraorbital canal and removal of a lateral and medial segment. To limit downward displacement of the eyeball, the anterior 1 cm of the orbital floor was left in place. For decompression of the lateral wall the ab interno approach (Paridaens et al., 2000) was used. This technique leaves the lateral rim intact, while minimally disinserting the temporalis muscle, thereby leading to less cosmetic disfigurement. This approach has the disadvantage of a limited visibility of the deep lateral wall (Kakizaki et al., 2011). Decompression of the lateral wall was performed as far posteriorly as reachable (approximately 1.5e2 cm), and up to the lacrimal key hole. 2.2. Clinical evaluation All patients underwent a thorough ophthalmic and endocrine investigation at the Multidisciplinary Orbital Clinic of University Medical Center Mainz, Germany preoperatively (1e3 days before surgery), as well as before discharge from hospital, and 3 months (12 ± 4 weeks) after surgery: Thyroid function and thyroid antibodies were evaluated by an endocrinologist with expertise in this field (GJK) in all patients. Ophthalmic examination included visual acuity testing, measurement of proptosis with the Hertel-exophthalmometer, slit lamp biomicroscopy, pupillary testing, and evaluation of clinical activity and severity scores as recommended by the European Group on Graves'Orbitopathy (EUGOGO) (Bartalena et al., 2008). A CT-scan was performed in all patients preoperatively to evaluate anatomical characteristics and to exclude maxillary sinusitis. The Gorman score was used to classify diplopia. In case of (suspected) dysthyroid optic neuropathy (DON) additional tests (vision field testing, measurement of the visual evoked potentials, color vision tests, MRI, and (if disk swelling was difficult to judge from fundoscopy alone) optical coherence tomography of the optic nerve head) were performed according to the recommendations of EUGOGO (Bartalena et al., 2008; Salvi et al., 2012) pre-, as well as postoperatively. In all cases of DON, high doses of intravenous glucocorticoids (in pulses of 500 mg methylprednisolone every second day; maximum cumulative dose: 8 g (McKeag et al., 2007; Zang et al., 2011) were
1815
administered first. The decision about the need for urgent decompression was made on the basis of the development of the clinical findings during therapy with intravenous steroids: Since DON may be accompanied by normal visual acuity, the decision was made not on visual acuity testing alone, but on the synopsis of all clinical signs, most importantly on fundoscopy, vision field testing, pupillary testing, visual evoked potentials, and color vision tests (McKeag et al., 2007). If the signs and symptoms of DON worsened or did not sufficiently improve after 1e2 weeks, decompression surgery was performed (Bartalena et al., 2008). In these cases, oral glucocorticoids were given after surgery and tapered down during the following 4 weeks. The main outcome measure was the median change of proptosis from preoperatively to postoperatively and to three months after surgery. Secondary outcome measure was the influence of patient characteristics (preoperative values of proptosis, age, gender, smoking status, level of specific autoantibodies, and whether patients had orbital radiotherapy prior to decompression surgery) on the surgical outcome. Furthermore, the long-term follow-up data were reviewed especially regarding changes of proptosis and the need for further surgeries during that period. 2.3. Statistical analysis Statistical analyses were performed using SPSS (Statistical Package for the Social Sciences, Version 21, Chicago, Illinois, USA), a commercially available software package. Statistical analyses were performed using SPSS (Statistical Package for the Social Sciences, Version 21, Chicago, Illinois, USA), a commercially available software package. The main question of this study was if the proptosis changed from preoperatively to postoperatively. To answer this question we applied a linear mixed model to account for the correlation structure between right and left eyes within one patient. Secondary questions (influence of age, gender, smoking status, level of specific autoantibodies and whether patients had orbital radiotherapy prior to decompression surgery on the change of proptosis from pre- to postoperatively or on the change of proptosis from postoperatively to 3 months after surgery) were analyzed in univariable linear mixed models. We fixed a significance level of 5% for the main question. All other analyses were performed exploratively. Because of the small number of cases having DON, a descriptive analysis was performed regarding the variables of optic neuropathy, only. 3. Results In 62 orbits of 40 consecutive and unselected patients, orbital decompression with the piezosurgical device was performed as a two (n ¼ 27, 44%) or three wall (35, 56%) decompression surgery. One patient (n ¼ 2 orbits, 3%) had additional blepharoplasty and two (n ¼ 4, 7%) had upper eyelid lengthening. 3.1. Patient characteristics Thirty-one (77.5%) were females and median age was 45.5 (range: 23e75) years. 22 (55%) patients were smokers. Graves' disease and Hashimoto's thyroiditis were present in 38 (95%) and one (3%) patients, respectively. No thyroid disease had been diagnosed in one (3%) patient, only. Median duration of thyroid disease at time of surgery was 62 (range: 13e640) months. At time of surgery, all patients were peripherally euthyroid. The levels of Thyroid stimulating Hormone-Receptor-Autoantibodies (TRAb) had been measured in 35 of 62 patients. Of those, 6 (16%) were TRAb-negative (24 mm) (Table 2). Furthermore, the difference of proptosis between 3 months and preoperatively was higher in females than in males (R ¼ 2.12 mm; 95% CI: 0.7e3.5 mm, p ¼ 0.005; univariable mixed linear model). There were no statistically significant effects of age, smoking status, level of thyroid stimulating hormone receptor antibodies, or previous retrobulbar irradiation on the change of proptosis.
Table 1 Preoperative ophthalmic findings of 40 consecutive patients with Thyroid associated orbitopathy. Proptosis, median (range, in mm) Visual Acuity, median (range, in logMAR) Severity of Thyroid associated orbitopathy (according to [3]), n (%) Mild Moderate-to-severe Sight-threatening (optic neuropathy) Diplopia (according to Gorman score [21]), n (%) No Intermittent Inconstant Constant
23 (16e30) 0.10 (0e1.3)
0 50 (81) 12 (19)
38 7 2 15
(61) (11) (3) (24)
Overall, visual acuity (in logMAR) increased by 0.43 (95% CI: 0.18e0.68) directly after and by 0.14 (95% CI: 0.15 to 0.42) 3 months after surgery (p ¼ 0.003; mixed linear model). An improvement of one and three Snellen lines was noted in 17 (27%) and 4 (7%) three months after surgery, whereas a deterioration was noted in 13 (21%) and 2 (3%) orbits, respectively. We looked in detail at the last mentioned cases with a deterioration of visual acuity of more than 3 Snellen lines: In one the reason was insufficient eyelid closure resulting in exposure keratopathy with photophobia and massive epiphora. Due to persisting proptosis, a second decompression surgery was planned to be performed in this patient. In the second case, the indication for surgery had been optic nerve compression. Other signs of optic nerve compression (afferent pupillary deficit, visual field) improved after surgery. In general, there might not be an improvement of visual acuity if there has been irreversible nerve fibre loss. In this certain case, there was another reason for progressive sight-loss: at 3 months, a development of cataract was noted. The high doses of intravenous steroids this patient had been given might have fastened cataract progression leading to a deterioration of visual acuity. In most cases, clinical signs of optic neuropathy improved after surgery: Optic disc crowding (in 2 orbits before surgery), pathologic visual evoked potentials (in 4 orbits before surgery), and color vision abnormalities (in 3 orbits before surgery) resolved in all cases after surgery, whereas in one orbit (of 6 orbits before surgery) a relative afferent pupillary deficit, and in 2 orbits (of 4 orbits preoperatively) vision field effects where still detectable 3 months after orbital decompression surgery. After three months, 4 of 24 (18%) patients experienced newonset diplopia, whereas diplopia had improved in 4 of 16 (22%). Information on the presence of infraorbital dys- or hypesthesia was available for 38 and 21 orbits directly after and at 3 months, respectively. It was present in 22 (58%) orbits directly after, and in 11 of 21 (52%) orbits at 3 months. An orbital floor lifting procedure had to be performed in 2 orbits during the follow-up as a significant hypoglobus had occurred. No case of postoperative hemorrhage, severe swelling and pain, or treatment requiring sinusitis were noted. 3.3.2. Long-term follow-up Long-term follow-up data were available of 53 orbits of 34 consecutive patients. The mean follow-up period was 14.6 ± 10.4 months (range: 6e42 months). Overall, mean proptosis decreased by further (from three months to long-term followup) 0.7 ± 2.0 mm (range: 7 to þ4; p < 0.001). In detail, proptosis had worsened in 4 orbits of 4 patients (in one patient within the first year after surgery and in 3 patients later on). In 2 (50%) of these cases the indication for surgery had been optic neuropathy. Furthermore, 3 (75%) of these patients were smokers. During the follow-up period 8 (15%) strabismus surgeries, 22 (42%) uppereyelid surgeries (blepharoplasty/eyelid lengthening), and 6 (11%) lower eyelid retraction surgeries were performed. 4. Discussion To the best our knowledge, the present study is the first to report original data to evaluate efficacy and safety of the piezo-instrument for bony orbital decompression surgery in TAO. We performed a detailed analysis of pre- and postoperative findings of a large collective of consecutive and unselected patients with different clinical degrees of TAO. An easy handling of the piezosurgical device, good visibility of the operation field, precise and selective cuttings during osteotomy, minimal intra-operative bleeding, minimal damage to the
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
1817
Fig. 2. a: Photograph of the use of the piezosurgical device for decompression of the orbital floor (1). View on the (intact) orbital floor. The handpiece is brought into position. b: Photograph of the use of the piezosurgical device for decompression of the orbital floor (2). Elevated bone fragment of the orbital floor (medially from the infraorbital canal).
surrounding soft tissue and a moderate postoperative swelling were appreciated by the surgeons. Complications of conventional decompression surgery include postoperative hemorrhages, severe swelling and pain, liquor leakage, mucocele, and painful traumatic neuroma (Mensink and Paridaens, 2006; Chang and Piva, 2008;
Baldeschi et al., 2010; Kakizaki et al., 2011; Patel et al., 2012). In the present study none of these complications were noted. There acquisition costs of piezosurgery are higher than those for the instruments for conventional surgery. Nevertheless, a better and faster healing after piezosurgery might be associated with
Fig. 3. a: Picture (of the orbit region and forehead) of a patient before orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). In this case, decompression surgery was performed for rehabilitative reasons as the patient suffered from disfiguring proptosis. b: Picture (from below; submentovertex) of a patient before orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). c: Picture (of the orbit region and forehead) of a patient three months after orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). After three-wall decompression surgery, exophthalmos was successfully cured. d: Picture (from below; submentovertex) of a patient after orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany).
1818
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
Fig. 4. Values of proptosis at baseline (preoperatively), postoperatively (before discharge from hospital) and 3 months after surgery. Error bars showing means ± standard errors of the mean. Proptosis was 23.6 ± 0.35 mm (mean ± standard error of the mean) at baseline and decreased to 20.8 ± 0.52 mm and to 20.9 ± 0.45 mm before discharge from hospital and 3 months after surgery, respectively (p < 0.001 according to mixed linear model).
shorter hospitalization times. This may lead to lower expenses over time. As disfiguring proptosis was the most frequent indication for surgery, we chose this parameter as mean outcome measure and benchmark of efficacy. A significant decrease of proptosis was noted especially in those with higher preoperative values. Furthermore, the effect after 3 months showed to be more successful in women. Nevertheless, this might have been due to the higher percentage of females in the collective (as women are more frequently affected by TAO). The other patient characteristics did not show any statistically significant influence on the reduction of proptosis after 3 months. Interestingly, almost all of the patients with a deterioration of proptosis during the long-term follow-up period were smokers. This might be associated with the fact that smoking is a risk factor for the development and response to therapies in TAO (Mann, 1999). Furthermore, in 50% of patients with a late-onset deterioration of proptosis, indication for surgery had been optic neuropathy and thus patients were operated within the active phase of the disease. At that point of time there was still an ongoing immune process within the orbit. This might have Table 2 Changes of proptosis in patients with higher preoperative values (>24 mm) versus in those with lower values (24 mm). The change of proptosis from preoperatively to postoperatively was higher in those with higher values (>24 mm) versus in those having had lower values (24 mm) before surgery: R ¼ 3.4 mm (95% CI: 4.4 to 2.6 mm) versus R ¼ 2.81 mm (95% CI: 3.6 to 2.0 mm). The same held true for the change of proptosis after 3 months: R ¼ 4.1 mm (95% CI: 4.9 to 3.3 mm) versus R ¼ 2.1 mm (95% CI: 2.8 to 1.4 mm; p < 0.001, respectively). Change of proptosis
From preoperatively to postoperatively From preoperatively to 3 months after surgery
R 95% CI p-value R 95% CI p-value
Proptosis 24 mm before surgery (N ¼ 38)
Proptosis > 24 mm before surgery (N ¼ 24)
2.8 mm 3.6 to 2.0 mm
Contents lists available at ScienceDirect
Journal of Cranio-Maxillo-Facial Surgery journal homepage: www.jcmfs.com
Piezosurgery for orbital decompression surgery in thyroid associated orbitopathy Katharina A. Ponto a, e, *, Isabella Zwiener b, e, Bilal Al-Nawas c, George J. Kahaly d, Anna F. Otto a, Julia Karbach c, Norbert Pfeiffer a, Susanne Pitz a a
Dept. of Ophthalmology (Head: Prof. Dr. N. Pfeiffer), University Medical Center Mainz, Germany Dept. of Biostatistics, University Medical Center Mainz, Germany Dept. of Maxillofacial Surgery, University Medical Center Mainz, Germany d Medicine I, University Medical Center Mainz, Germany e Center of Thrombosis and Hemostasis (CTH), University Medical Center Mainz, Germany b c
a r t i c l e i n f o
a b s t r a c t
Article history: Paper received 9 January 2014 Accepted 18 June 2014 Available online 26 June 2014
The purpose of this study was to assess a piezosurgical device as a novel tool for bony orbital decompression surgery. At a multidisciplinary orbital center, 62 surgeries were performed in 40 patients with thyroid associated orbitopathy (TAO). Within this retrospective case-series, we analyzed the medical records of these consecutive unselected patients. The reduction of proptosis was the main outcome measure. Indications for a two (n ¼ 27, 44%) or three wall (35, 56%) decompression surgery were proptosis (n ¼ 50 orbits, 81%) and optic neuropathy (n ¼ 12, 19%). Piezosurgery enabled precise bone cuts without intraoperative complications. Proptosis decreased from 23.6 ± 2.8 mm (SD) by 3 mm (95% CI: 3.6 to 2.5 mm) after surgery and stayed stable at 3 months (3 mm, 95% CI: 3.61 to 2.5 mm, p < 0.001, respectively). The effect was higher in those with preoperatively higher values (>24 mm versus 24 mm: 3.4 mm versus 2.81 mm before discharge from hospital and 4.1 mm versus 2.1 mm at 3 months: p < 0.001, respectively). After a mean long-term follow-up period of 14.6 ± 10.4 months proptosis decreased by further 0.7 ± 2.0 mm (p < 0.001). Signs of optic nerve compression improved after surgery. Infraorbital hypesthesia was present in 11 of 21 (52%) orbits 3 months after surgery. The piezosurgical device is a useful tool for orbital decompression surgery in TAO. By cutting bone selectively, it is precise and reduces the invasiveness of surgery. Nevertheless, no improvement in outcome or reduction in morbidity over conventional techniques has been shown so far. © 2014 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
Keywords: Orbital surgery Decompression Thyroid associated orbitopathy Piezosurgery Proptosis
1. Introduction Thyroid associated orbitopathy (TAO) is the most frequent extrathyroidal manifestation of autoimmune hyperthyroidism (Bahn, 2010). In the inflammatory active phase of the disease, immunomodulatory treatment and/or retrobulbar irradiation are indicated (Chang and Piva, 2008; Douglas and Gupta, 2011; Kakizaki et al., 2011). In patients with severe but inactive TAO and/or in optic neuropathy orbital decompression surgery is
* Corresponding address. Dept. of Ophthalmology, University Medical Center Mainz, Langenbeckstr. 1, 55131 Mainz, Germany. Tel./fax: þ49 06131/17 7085x2290. E-mail addresses: [email protected], [email protected] (K.A. Ponto).
required (Bartalena et al., 2008; Baldeschi et al., 2010; Patel et al., 2012). In addition, decompression surgery may also be performed for prolonged orbital congestion, pain, or for cases with severe corneal exposure due to insufficient eyelid closure. As quality of life is markedly impaired in patients with severe proptosis (Robiony and Polini, 2010; Cassetta et al., 2012), cosmesis is increasingly recognized as an indication for orbital decompression. Nowadays, the swinging-eyelid approach has become the most popular method followed by the coronal and transconjunctival approaches (Paridaens et al., 2006; Baldeschi et al., 2010; Ponto et al., 2011, 2013; Zang et al., 2011). Nevertheless, in conventional surgery performed with hammer and chisel, damage to adjacent soft tissue is difficult to avoid, and the resulting bone fragments are barely predictable. Clinical studies especially from maxillofacial-, dental-, ENT- and neurosurgery have shown benefits of ultrasonic bone
http://dx.doi.org/10.1016/j.jcms.2014.06.020 1010-5182/© 2014 European Association for Cranio-Maxillo-Facial Surgery. Published by Elsevier Ltd. All rights reserved.
1814
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
cutting (Lytton et al., 2010; Ponto and Kahaly, 2010; Ponto et al., 2010; Zang et al., 2011). So far, studies on the piezosurgical device for orbital surgery are sparse and original data on efficacy and safety are lacking (Ponto et al., 2009). Due to the saw-like cutting properties modifications of the osteotomy procedure have to be established. We therefore aimed to evaluate this technique for bony orbital decompression surgery in patients with TAO at our institution. At the multidisciplinary orbital center of University Medical Center, Mainz, Germany, 62 bony orbital decompression surgeries with the piezosurgical device were performed in 40 patients with TAO between February 2010 and March 2012. Within this retrospective case-series, we analyzed he medical records or files of these consecutive unselected patients. The protocol was consistent with the principles of the Declaration of Helsinki. Because this retrospective case-series did not include any interventions aside from those commonly falling within the daily routine and because none of the individual-related data were passed to third parties, the responsible Ethics Committee of the Medical Association of the State Rhineland Palatinate, Germany, decided that no approval was required. 2. Material and methods 2.1. Surgical procedure Piezoelectric bone surgery (syn. Piezosurgery), is a technique of osteotomy that requires microvibrations of ultrasonic frequency scalpels. The principle of piezosurgery is ultrasonic transduction, obtained by piezoelectric ceramic contraction and expansion. The resulting vibrations are amplified and transferred onto the insert of a tip which results in the cavitation phenomenon with a
mechanical cutting effect, exclusively on mineralized tissues: with the piezosurgical device vibrating at 25e29 kHz, bones are cut selectively while sparing fine anatomical structures (e.g., vessels, nerve tissue), that would only be damaged when using higher frequencies. An oscillating tip with irrigation fluid enables effective cooling as well as higher visibility (via cavitation effect) (Schlee et al., 2006; Crosetti et al., 2009). The piezosurgical device we used (mectron™ piezosurgery, Cologne, Germany) consists of a basis unit, a surgical handpiece, and a footswitch. Irrigation and aspiration occur at the handpiece tip. For orbital decompression surgery, we used the saw-like tip (OT7) for ultrasonic bone cutting. The amplitude of tip vibration, rate of irrigation, and aspiration pressure can be regulated on the basis unit. A sterile saline solution is used for irrigation and cooling. A rate between 3 and 40 ml/min is used for irrigation. Before clinical use, the handling of the instrument for orbital decompression surgery was practiced on corpses of the institute of anatomy, Gutenberg University, Mainz, Germany (Fig. 1aed). The surgical procedure was personalized for each patient depending on the degree of proptosis, the presence of optic neuropathy, and the presence of eye motility disturbances. As a general principle, the decision about a two or three wall decompression surgery based on clinical severity: In the presence of optic neuropathy and high values of proptosis (cut-off: 22 mm) the effect of inferomedial decompression was implemented by adding lateral wall decompression (foremost of the inferior lateral orbital quadrant). We chose the relatively low cut-off of 22 mm because this value had been defined as a cut-off to dose the amount of decompression before (Paridaens et al., 2006) and because we aimed to perform a balanced decompression in most patients. Balanced orbital decompression (medial and lateral orbital wall with or without decompression of the orbital floor) has been
Fig. 1. Photographs at a practice session at the Institute of Anatomy. a: Exposition of the orbital floor via a transconjunctival approach with lateral canthotomy (“swinging eyelid”). b: Osteotomy with the piezosurgical device (OT7 tip). c: Bone removal laterally from the infraorbital nerve; an osteotomy medially from the nerve has already been performed. d: Removal of the medial segment of the orbital floor; the region of the infraorbital nerve is spared.
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
shown to be beneficial regarding the development of postoperative diplopia and motility disturbances (Goldberg et al., 2000; Unal et al., 2000; Kacker et al., 2003; Takahashi et al., 2008; Barkhuysen et al., 2009; Alsuhaibani et al., 2011). In patients with less proptosis, the lateral wall was not removed to avoid overcorrection (enophthalmos). All decompression surgeries were performed via a combined inferior fornix and transcaruncular approach. After conjunctival incision, the periosteum was incised along the inferior and lateral orbital rim. The periorbita was separated bluntly from the medial, inferior, and lateral orbital walls and incised in a posterioreanterior direction medially, infero-medially, and laterally. Approximately 1e1.5 cm of the anterior periorbita was left intact and herniation of fat was augmented through blunt dissection with scissors in the spaces between the rectus muscles. For decompression of the medial orbital wall we chose the transcaruncular approach as it is a valid alternative to the endoscopic approach (Shorr et al., 2000; Liao et al., 2006). It has been shown that decompression of the medial orbital wall successfully relieves the pressure at the orbital apex (McCann et al., 2006). The ethmoidal part of the medial orbital wall was decompressed as far posteriorly as reachable (approximately 1.5e2 cm) and as felt necessary for proptosis reduction. At our institution, we have gained positive experience with this method, even though it might be more difficult to quantify the amount of bone removal as it is possible under direct visualization using the endoscopic approach. Bone of the orbital floor was removed in two parts by cutting along the infraorbital canal and removal of a lateral and medial segment. To limit downward displacement of the eyeball, the anterior 1 cm of the orbital floor was left in place. For decompression of the lateral wall the ab interno approach (Paridaens et al., 2000) was used. This technique leaves the lateral rim intact, while minimally disinserting the temporalis muscle, thereby leading to less cosmetic disfigurement. This approach has the disadvantage of a limited visibility of the deep lateral wall (Kakizaki et al., 2011). Decompression of the lateral wall was performed as far posteriorly as reachable (approximately 1.5e2 cm), and up to the lacrimal key hole. 2.2. Clinical evaluation All patients underwent a thorough ophthalmic and endocrine investigation at the Multidisciplinary Orbital Clinic of University Medical Center Mainz, Germany preoperatively (1e3 days before surgery), as well as before discharge from hospital, and 3 months (12 ± 4 weeks) after surgery: Thyroid function and thyroid antibodies were evaluated by an endocrinologist with expertise in this field (GJK) in all patients. Ophthalmic examination included visual acuity testing, measurement of proptosis with the Hertel-exophthalmometer, slit lamp biomicroscopy, pupillary testing, and evaluation of clinical activity and severity scores as recommended by the European Group on Graves'Orbitopathy (EUGOGO) (Bartalena et al., 2008). A CT-scan was performed in all patients preoperatively to evaluate anatomical characteristics and to exclude maxillary sinusitis. The Gorman score was used to classify diplopia. In case of (suspected) dysthyroid optic neuropathy (DON) additional tests (vision field testing, measurement of the visual evoked potentials, color vision tests, MRI, and (if disk swelling was difficult to judge from fundoscopy alone) optical coherence tomography of the optic nerve head) were performed according to the recommendations of EUGOGO (Bartalena et al., 2008; Salvi et al., 2012) pre-, as well as postoperatively. In all cases of DON, high doses of intravenous glucocorticoids (in pulses of 500 mg methylprednisolone every second day; maximum cumulative dose: 8 g (McKeag et al., 2007; Zang et al., 2011) were
1815
administered first. The decision about the need for urgent decompression was made on the basis of the development of the clinical findings during therapy with intravenous steroids: Since DON may be accompanied by normal visual acuity, the decision was made not on visual acuity testing alone, but on the synopsis of all clinical signs, most importantly on fundoscopy, vision field testing, pupillary testing, visual evoked potentials, and color vision tests (McKeag et al., 2007). If the signs and symptoms of DON worsened or did not sufficiently improve after 1e2 weeks, decompression surgery was performed (Bartalena et al., 2008). In these cases, oral glucocorticoids were given after surgery and tapered down during the following 4 weeks. The main outcome measure was the median change of proptosis from preoperatively to postoperatively and to three months after surgery. Secondary outcome measure was the influence of patient characteristics (preoperative values of proptosis, age, gender, smoking status, level of specific autoantibodies, and whether patients had orbital radiotherapy prior to decompression surgery) on the surgical outcome. Furthermore, the long-term follow-up data were reviewed especially regarding changes of proptosis and the need for further surgeries during that period. 2.3. Statistical analysis Statistical analyses were performed using SPSS (Statistical Package for the Social Sciences, Version 21, Chicago, Illinois, USA), a commercially available software package. Statistical analyses were performed using SPSS (Statistical Package for the Social Sciences, Version 21, Chicago, Illinois, USA), a commercially available software package. The main question of this study was if the proptosis changed from preoperatively to postoperatively. To answer this question we applied a linear mixed model to account for the correlation structure between right and left eyes within one patient. Secondary questions (influence of age, gender, smoking status, level of specific autoantibodies and whether patients had orbital radiotherapy prior to decompression surgery on the change of proptosis from pre- to postoperatively or on the change of proptosis from postoperatively to 3 months after surgery) were analyzed in univariable linear mixed models. We fixed a significance level of 5% for the main question. All other analyses were performed exploratively. Because of the small number of cases having DON, a descriptive analysis was performed regarding the variables of optic neuropathy, only. 3. Results In 62 orbits of 40 consecutive and unselected patients, orbital decompression with the piezosurgical device was performed as a two (n ¼ 27, 44%) or three wall (35, 56%) decompression surgery. One patient (n ¼ 2 orbits, 3%) had additional blepharoplasty and two (n ¼ 4, 7%) had upper eyelid lengthening. 3.1. Patient characteristics Thirty-one (77.5%) were females and median age was 45.5 (range: 23e75) years. 22 (55%) patients were smokers. Graves' disease and Hashimoto's thyroiditis were present in 38 (95%) and one (3%) patients, respectively. No thyroid disease had been diagnosed in one (3%) patient, only. Median duration of thyroid disease at time of surgery was 62 (range: 13e640) months. At time of surgery, all patients were peripherally euthyroid. The levels of Thyroid stimulating Hormone-Receptor-Autoantibodies (TRAb) had been measured in 35 of 62 patients. Of those, 6 (16%) were TRAb-negative (24 mm) (Table 2). Furthermore, the difference of proptosis between 3 months and preoperatively was higher in females than in males (R ¼ 2.12 mm; 95% CI: 0.7e3.5 mm, p ¼ 0.005; univariable mixed linear model). There were no statistically significant effects of age, smoking status, level of thyroid stimulating hormone receptor antibodies, or previous retrobulbar irradiation on the change of proptosis.
Table 1 Preoperative ophthalmic findings of 40 consecutive patients with Thyroid associated orbitopathy. Proptosis, median (range, in mm) Visual Acuity, median (range, in logMAR) Severity of Thyroid associated orbitopathy (according to [3]), n (%) Mild Moderate-to-severe Sight-threatening (optic neuropathy) Diplopia (according to Gorman score [21]), n (%) No Intermittent Inconstant Constant
23 (16e30) 0.10 (0e1.3)
0 50 (81) 12 (19)
38 7 2 15
(61) (11) (3) (24)
Overall, visual acuity (in logMAR) increased by 0.43 (95% CI: 0.18e0.68) directly after and by 0.14 (95% CI: 0.15 to 0.42) 3 months after surgery (p ¼ 0.003; mixed linear model). An improvement of one and three Snellen lines was noted in 17 (27%) and 4 (7%) three months after surgery, whereas a deterioration was noted in 13 (21%) and 2 (3%) orbits, respectively. We looked in detail at the last mentioned cases with a deterioration of visual acuity of more than 3 Snellen lines: In one the reason was insufficient eyelid closure resulting in exposure keratopathy with photophobia and massive epiphora. Due to persisting proptosis, a second decompression surgery was planned to be performed in this patient. In the second case, the indication for surgery had been optic nerve compression. Other signs of optic nerve compression (afferent pupillary deficit, visual field) improved after surgery. In general, there might not be an improvement of visual acuity if there has been irreversible nerve fibre loss. In this certain case, there was another reason for progressive sight-loss: at 3 months, a development of cataract was noted. The high doses of intravenous steroids this patient had been given might have fastened cataract progression leading to a deterioration of visual acuity. In most cases, clinical signs of optic neuropathy improved after surgery: Optic disc crowding (in 2 orbits before surgery), pathologic visual evoked potentials (in 4 orbits before surgery), and color vision abnormalities (in 3 orbits before surgery) resolved in all cases after surgery, whereas in one orbit (of 6 orbits before surgery) a relative afferent pupillary deficit, and in 2 orbits (of 4 orbits preoperatively) vision field effects where still detectable 3 months after orbital decompression surgery. After three months, 4 of 24 (18%) patients experienced newonset diplopia, whereas diplopia had improved in 4 of 16 (22%). Information on the presence of infraorbital dys- or hypesthesia was available for 38 and 21 orbits directly after and at 3 months, respectively. It was present in 22 (58%) orbits directly after, and in 11 of 21 (52%) orbits at 3 months. An orbital floor lifting procedure had to be performed in 2 orbits during the follow-up as a significant hypoglobus had occurred. No case of postoperative hemorrhage, severe swelling and pain, or treatment requiring sinusitis were noted. 3.3.2. Long-term follow-up Long-term follow-up data were available of 53 orbits of 34 consecutive patients. The mean follow-up period was 14.6 ± 10.4 months (range: 6e42 months). Overall, mean proptosis decreased by further (from three months to long-term followup) 0.7 ± 2.0 mm (range: 7 to þ4; p < 0.001). In detail, proptosis had worsened in 4 orbits of 4 patients (in one patient within the first year after surgery and in 3 patients later on). In 2 (50%) of these cases the indication for surgery had been optic neuropathy. Furthermore, 3 (75%) of these patients were smokers. During the follow-up period 8 (15%) strabismus surgeries, 22 (42%) uppereyelid surgeries (blepharoplasty/eyelid lengthening), and 6 (11%) lower eyelid retraction surgeries were performed. 4. Discussion To the best our knowledge, the present study is the first to report original data to evaluate efficacy and safety of the piezo-instrument for bony orbital decompression surgery in TAO. We performed a detailed analysis of pre- and postoperative findings of a large collective of consecutive and unselected patients with different clinical degrees of TAO. An easy handling of the piezosurgical device, good visibility of the operation field, precise and selective cuttings during osteotomy, minimal intra-operative bleeding, minimal damage to the
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
1817
Fig. 2. a: Photograph of the use of the piezosurgical device for decompression of the orbital floor (1). View on the (intact) orbital floor. The handpiece is brought into position. b: Photograph of the use of the piezosurgical device for decompression of the orbital floor (2). Elevated bone fragment of the orbital floor (medially from the infraorbital canal).
surrounding soft tissue and a moderate postoperative swelling were appreciated by the surgeons. Complications of conventional decompression surgery include postoperative hemorrhages, severe swelling and pain, liquor leakage, mucocele, and painful traumatic neuroma (Mensink and Paridaens, 2006; Chang and Piva, 2008;
Baldeschi et al., 2010; Kakizaki et al., 2011; Patel et al., 2012). In the present study none of these complications were noted. There acquisition costs of piezosurgery are higher than those for the instruments for conventional surgery. Nevertheless, a better and faster healing after piezosurgery might be associated with
Fig. 3. a: Picture (of the orbit region and forehead) of a patient before orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). In this case, decompression surgery was performed for rehabilitative reasons as the patient suffered from disfiguring proptosis. b: Picture (from below; submentovertex) of a patient before orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). c: Picture (of the orbit region and forehead) of a patient three months after orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany). After three-wall decompression surgery, exophthalmos was successfully cured. d: Picture (from below; submentovertex) of a patient after orbital decompression surgery. Three-dimensional image; photo-optical scanner (Activity 102; Smart Optics, Bochum, Germany).
1818
K.A. Ponto et al. / Journal of Cranio-Maxillo-Facial Surgery 42 (2014) 1813e1820
Fig. 4. Values of proptosis at baseline (preoperatively), postoperatively (before discharge from hospital) and 3 months after surgery. Error bars showing means ± standard errors of the mean. Proptosis was 23.6 ± 0.35 mm (mean ± standard error of the mean) at baseline and decreased to 20.8 ± 0.52 mm and to 20.9 ± 0.45 mm before discharge from hospital and 3 months after surgery, respectively (p < 0.001 according to mixed linear model).
shorter hospitalization times. This may lead to lower expenses over time. As disfiguring proptosis was the most frequent indication for surgery, we chose this parameter as mean outcome measure and benchmark of efficacy. A significant decrease of proptosis was noted especially in those with higher preoperative values. Furthermore, the effect after 3 months showed to be more successful in women. Nevertheless, this might have been due to the higher percentage of females in the collective (as women are more frequently affected by TAO). The other patient characteristics did not show any statistically significant influence on the reduction of proptosis after 3 months. Interestingly, almost all of the patients with a deterioration of proptosis during the long-term follow-up period were smokers. This might be associated with the fact that smoking is a risk factor for the development and response to therapies in TAO (Mann, 1999). Furthermore, in 50% of patients with a late-onset deterioration of proptosis, indication for surgery had been optic neuropathy and thus patients were operated within the active phase of the disease. At that point of time there was still an ongoing immune process within the orbit. This might have Table 2 Changes of proptosis in patients with higher preoperative values (>24 mm) versus in those with lower values (24 mm). The change of proptosis from preoperatively to postoperatively was higher in those with higher values (>24 mm) versus in those having had lower values (24 mm) before surgery: R ¼ 3.4 mm (95% CI: 4.4 to 2.6 mm) versus R ¼ 2.81 mm (95% CI: 3.6 to 2.0 mm). The same held true for the change of proptosis after 3 months: R ¼ 4.1 mm (95% CI: 4.9 to 3.3 mm) versus R ¼ 2.1 mm (95% CI: 2.8 to 1.4 mm; p < 0.001, respectively). Change of proptosis
From preoperatively to postoperatively From preoperatively to 3 months after surgery
R 95% CI p-value R 95% CI p-value
Proptosis 24 mm before surgery (N ¼ 38)
Proptosis > 24 mm before surgery (N ¼ 24)
2.8 mm 3.6 to 2.0 mm
