The isolated orbital floor fracture from a transconjunctival or subciliary perspective-A standardized anthropometric evaluation

Background The influence of orbital fractures and their repair on the rate of deformities of the lower eyelid is an ongoing source of discussion in the literature. Most of the present studies include isolated blowout as well as combined orbital fractures. Material and Methods We present a retrospective evaluation of a series of 100 patients after isolated blowout fracture repair using reference anthropometric data on standardized photographs. Analysis included eye fissure width and height, lid sulcus height, upper lid height, upper and lower iris coverage, position of cornea to palpebra inferior, canthal tilt, scleral show, ectropion and entropion. It was clearly distinguished between operated and contralateral eyelid, whether a transconjunctival or a subciliary approach was performed and amount of fracture. Our main interests were changes of the aforementioned parameters with regards to eyelid deformities. Results Surgery per se did not significantly influence eyelid deformities. However, the surgical approach selected significantly affected eye fissure index, lower iris coverage and rate of scleral show, indicating retraction of the lower eyelid. Conclusions The standardized measurements described here are accurate and objective to evaluate postoperative results. The subciliary approach included the highest risk of lower lid retraction as compared to transconjunctival approaches. Key words:Transconjunctical approach, subciliary approach, orbital floor fracture.


Introduction
Accordin� to our e�perience, before under�oin� sur�ical repair of a blo� out fracture, �ost patients �orry about the ris� of distortion of the face and especially the eyelids. Even �ini�ally displaced blo�-out fractures �ay result in aesthetic and functional defor�ities of the periorbital re�ion (1). There is an on�oin� discussion in the literature about the optimal treatment of orbital floor fractures. Especially the discussion about how to approach the orbital floor is on�oin�. To date �ost studies co�parin� transconjunctival and transcutaneous approaches include patients with isolated orbital floor fractures, zygomaticomaxillary fractures and co�bined orbito�a�illary fractures alto�ether (2,�) �ithout �ivin� results clearly distin-�uishin� bet�een these different entities of fractures. �t see�s reasonable, as reported earlier, that different severity and type of trauma have significant impact on the ris� of developin� an en-or ectropion (�). Thus the inclusion of different types of fractures of the orbita in studies referrrin� and/or co�parin� transcutaneous and transconjunctival approaches li�its their validity. Only fe� articles referrin� to a sin�le type of fracture are available. These articles �ostly report on the out-co�e of isolated blo�-out fractures (4)(5)(6)(7). They report the clinical �ana�e�ent (7), functional outco�e (�, 6) and clinical outco�e of the sur�ical �ethod (4,6). There is a lac� of elaborated and objective assess�ents of the effect of blo� out fractures and its sur�ical treat�ent on the eyelid architecture in the current literature. �o�ever, such an assess�ent is hi�hly desirable, as it may help to quantify the influence of trauma and particular sur�ical procedure selected on the eyelid �orphol-o�y. Nor�ative anthropo�etric �easure�ents of the face are available (8)(9)(10)(11)(12)(13). Their benefit in planning, per-for�ance and evaluation of facial sur�ery is �idely rec-o�nized (11,12,14). �n a �roup of 100 patients sufferin� fro� isolated blo�-out fractures, anthropo�etric �eas-ure�ents �ere perfor�ed on standardized photo�raphs. We investi�ated differences bet�een the affected and the contralateral side and either a transconjunctival or a subciliary approach �as perfor�ed. �urther�ore �e evaluated the influence of the type of orbital floor fracture.

Material and Methods
Before the study �as initiated, the local Ethics Co��ittee of the �niversity �ospital Jena �as as�ed to �ive his approval to the study. Because the study desi�n ai�ed to evaluate routinely perfor�ed docu�entation li�e standardized photographies or X-rays and did not influence the the dia�nostical or therapeutic process the Ethics Co��ittee denied the necessity of special ethical approval. Prior to sur�ery all included patients si�ned an informed consent permitting the scientific evaluation of their routinely recorded docu�entation includin� �rays and photo�raphies. All patients �ere operated at the Depart�ent of Plastic Sur�ery � Cranio-Ma�illofacial Sur�ery at the �niversity �ospital Jena, Ger�any, bet�een January 2006 and Dece�ber 2011. The inferior orbital ri� and orbital floor were exposed either through a subciliary or a transconjunctival approach, �hich �ere perfor�ed in a standardized �anner. The subciliary approach �as perfor�ed in the �anner of a step dissection, the transconjunctival approach in a retroseptal techni�ue. A photo-and radio�raphic description of three patients is sho�n in fi�ure 1. Colored frontal vie� photo�raphs �ith open eyes �ere taken postoperatively, after definite woundhealing, with a Ni�on D 80 ca�era (objective: Ni�on A� Micro Ni�-�or 10� �� 1:2.8 D; aperture: f1�; Ni�on Corp, To�yo, Japan) �ith a standardized lens at a patient distance of 1 � in a standardized position and a slit la�p by a professional photo�rapher. Only photo�raphs in �hich the interpupillary a�is �as at the sa�e level as the ca�era lens and faces �ere clearly at rest �ere selected to �ini-�ize photo�raphic distortion (1�,16). �urther analysis �as perfor�ed usin� Adobe Photoshop CS2 (Adobe �nc, San Jose, CA). On the basis of predefined landmarks and data ( Table  1), the follo�in� anthropo�etric di�ensions based on the �or� of �ar�as and Munro (9)(10)(11)(12)14) as �ell as �ell �no�n clinical data �ere investi�ated (�i�. 2): Eye �issure Index is defined by the eye fissure height (EFH, Ps-Pi), the vertical distance fro� the �ar�in of the inferior palpebra to the �ar�in of the superior palpebra. The EFH was then divided by the eye fissure width (EFW, en-ex), which is defined by the intercanthal distance. The eyelid sulcus of the upper eyelid divides the upper eyelid in an upper and lo�er part. The upper lid sulcus hei�ht (�LS�, LS-Ps) is depicted by the vertical dis-tance bet�een the upper palpebral �ar�in and eyelid sulcus. as percenta�e of the upper lid hei�ht (�L�, Os-Ps), the distance bet�een orbitale superioris and upper palpebral �ar�in. �pper iris covera�e (��C) represents the part of the upper iris covered by the upper eyelid. �t �as investi�ated by halvin� iris dia�eter and subtract-in� the free visible upper radius of the iris (�c-Ps) as percenta�e of the total iris dia�eter (�D). Lo�er iris covera�e (L�C) represents the part of the lo�er iris covered by the lo�er eyelid. �t �as raised by halvin� the iris dia�eter and subtractin� the free visible upper radius of the iris (�c-Pi). �n the case of scleral sho� or ectropion its values turned ne�ative. The position of the lo�er eyelid to the lo�er iris describes the an�ulation of the inferior eyelid to the center of the iris (8). �t �as �easured by placin� a vertical reference line throu�h the center of the iris (�c). Another line �as dra�n throu�h the center of the iris (�c) and the point of contact of the lo�er eyelid and cornea (�c-CPi). The an�le for�ed by both lines �as �easured in de�rees (�i�. �). Medial deviations of the an�le �ere �easured as ne�ative, lateral deviations as positive value. Canthal tilt describes the intercanthal fissure inclination (13) measured as the angle between the E�W (en-e�) and a horizontal reference line passin� throu�h the endocanthion in de�rees (�i�. �). �urther-�ore the rate of scleral sho�, ectropion, and entropion �as recorded. All para�eters �ere �easured on both eyes. �esults �ere evaluated co�parin� the operated and the contralateral (not operated, control) side. The i�pact of �hether a transconjunctival or a subciliary approach �as perfor�ed �as evaluated, as �ell. �urther�ore the influence of the type of orbital floor fracture was in-vesti�ated throu�h an analysis of operation reports and preoperative CT scans �ith coronal and sa�ittal refor-�ations. Type 1 consisted of s�all fractures of the anterior medial orbital floor and type 2 of larger fractures involving the orbital floor and medial wall (17). Occurrence of diplopia �as e�tracted out of patients´ records.
In order to analyse the influence of operated and contralateral side, sur�ical approach selected and type of orbital floor fracture on EFI, ULSH, UIC, LIC, position of lo�er eyelid to lo�er iris and canthal tilt, univariate and �i�ed �odel (   approach selected. One ectropion �as observed in the �roup of a subciliary approach. The type of orbital floor fracture did not significantly influence on the parameters investigated. The �ultivariate analysis perfor�ed did not yield si�nificant interaction effects between the factors operated or not, sur�ical approach and type and severity of fracture. However, for statistical reasons a significant interaction effect is not required to confirm the significant effect of the sur�ical approach on E��, L�C and scleral sho� values. T�o patients (2.2%) suffered fro� persistent diplopia in the direction of ocular elevation at the ti�e the postoperative photo�raphs �ere ta�en. Both patients under-�ent a transconjunctival approach. None of the� presented sy�pto�s of entrap�ent or enophtal�os in the postoperative ophthal�olo�ic e�a�ination. We �ere unable to find a medical, anatomic or surgical reason, �hich is not unusual (1�).

Discussion
A blow-out fracture is defined as a fracture of the orbital floor. It does not involve the orbital rim. Besides the description of functional disabilities the �ost co��on criteria of postoperative evaluation of orbital floor fracture repair consists in the rate of lo�er lid retraction, ectropion and entropion (4). These co��on criteria do not allo� detection of �ore subtle chan�es of the periorbital architecture. The presented anthropo�etric �easure�ents of the periorbital re�ion �ay help us to objectify the �orphologic outcome of orbital floor fracture repair. As different �rades of severity and types of trau�a play a decisive role in the ris� of develop�ent of en-or ectropion (2,�), �e included only isolated blo�-out fractures in our study, to i�prove the validity of our data. The si�nificance of the investigation of the impact of subciliary or transconjunctival approaches on the periorbital architecture are enhanced thereby, as �ell.
Orbital floor fractures result from an abrupt increase of intraorbital pressure and �ay be caused either by direct contact to the �lobe or contact �ith the inferior orbital rim causing the floor to buckle. Forces applied to the orbital ri�, described by Waterhouse et al. as type 1, rather lead to small fractures of the mid medial floor and rarely herniation of orbital content. �orces applied to the �lobe rather lead to lar�er fractures includin� the orbital floor and medial wall and herniation of orbital content and �ere described by Waterhouse et al. as type 2 (17). Due to the potential influence of the type of fracture to postoperative eyelid �alposition this easy and reproducible classification was used to investigate the influence of amount of fracture on eyelid morphology. Several anthropo�etric �easure�ents of the periorbital re�ion have been described (9,10,12,14). We used the eye fissure index (EFI), upper lid sulcus height (ULSH), upper (��C) and lo�er (L�C) iris covera�e, canthal tilt and position of lo�er eyelid to iris in our study. The eye fissure width, measured between the endo-and exocanthion, is referred to equal 30 mm. The eye fissure hei�ht bet�een the �ar�ins of the upper and lo�er palpebra is reported to be 9-10 �� �ith open eyes strai�ht ahead (18). Because linear �easure�ents are not e�actly reproducible in standardized photo�raphs, we preferred to apply the EFI reflecting the relation be-t�een E�� and E�W. The L�C is very i�portant for the loo� of the patient. The nor�ative value is 7% (12). Ne�ative values occur in the case of scleral sho�. Sclera should nor�ally not be visible loo�in� strai�ht ahead (8). A reproducible photographic quantification of scleral show is desirable for the jud�e�ent of the �uantity of distortion. Therefore scleral show was quantified by changes of EFI and L�C.
Ectropia are lin�ed to lo�er lid retraction, as �ell, but not in such a direct �anner as scleral sho�. Scleral sho� describes a �eneral and sy��etric decline of the lo�er eyelid attached to the eye�lobe. �n case of an ectropion the lo�er eyelid turns inside out, leavin� the inner eyelid and �lobe surface e�posed and is subse�uently prone to irritation. �t �ay occur �edially or laterally or on both sides and does not inevitably �o alon� �ith e�cessive lo�er lid retraction. Measure�ents of the upper eyelid position �ere included in our study in order to secure that chan�es of the �orpholo�y of the upper eyelid did not affect the �easure�ents of E��. �LS� is a helpful �easure�ent in the appraisal of the co�position of the eyelid to the eyebrow. UIC reflects the covered part of the upper iris (12).
To ade�uately describe the shape of the eyelids t�o an�les e�hibitin� decisive i�pact on the periorbital appearance �ere �easured: Canthal tilt (1�) is of bi� concern for the facial appearance. Sad loo� �ay be the conse�uence of a ne�ative canthal tilt (8). �t �as referred to be 2 �� or at an an�le of 10 to 1� de�rees above the �edial canthus (19). The position of lo�er eyelid relative to iris describes the nor�al contact point of the lo�er palpebra to the li�bus corneae at the 6 o´cloc� position (8).
Clearly identifiable eyelid distortions such as unilateral lo�er lid retraction and scleral sho� or a lo�ered canthal tilt lead to an unpleaseant appearance, �hich often is noticed by the patients the�selves.
Alto�ether the nine presented anthropo�etric and clinically relevant para�eters described in this study are able to describe and �uantify such �alpositions. They were easily and reproduciblely definable in the frontal view photographs and may be influenced by a blow-out fracture or its sur�ical repair. The co�parison of post-operative photo�raphs by sur�eons and/or independent observers see�s less reproducible to us than the presented anthropo�etric �easure�ents. The consideration of the anthropo�etric para�eters described may be relevant not only for scientific purposes but also in the clinical care of these patients. �f in the further clinical course a sur�ical revision is �arranted, it is i�portant to e�actly plan the de�ree of correction necessary. �n order to achieve the best result possible it is not only necessary to e�actly esti�ate the de�ree of vertical correction described in this study by E�� and L�C, but also to achieve an appealin� shape of the lo�er eyelid to�ards the �lobe. Canthal tilt and position of lo�er eyelid to iris �ay facilitate this esti�ation. �n the presented study �e ai�ed to focus on �orphologic aspects and the influence of trauma and surgical approach. Previous studies indicated, that the interpretation of the raw data of ophthalmologic findings do not correlate �ith the "real life" rate of co�plications. Therefore the ophthal�olo�ic evaluation has to be interpreted for every individual patient and �as not evaluated and discussed in detail in this current study (�). The co�parison of operated and contralateral side as well as of the surgical approach to the orbital floor did not exhibit a significant effect on ULSH, UIC, canthal tilt and position of lo�er eyelid to iris (see Table 2). The constant values of ��C and �LS� indicate that, not sur-prisin�ly, the architecture of the upper eyelid and the shape of the eyelids were not influenced by the blow-out fracture and its subse�uent repair. EFI and LIC did not show significant differences, when operated and contralateral side �ere co�pared (see Table 2). This underlines, that pree�istin� scleral sho� on one side, �hich is often associated �ith scleral sho� on the contralateral side, has no significant influence on the rate of postoperative scleral sho�. �urther�ore it could be interpreted as an indication, that sur�ery itself is not associated �ith hi�her rates and a�ount of eyelid defor�ities. �o�ever increased values of E��, decreased values of L�C and an increased rate of sleral sho� �ere observed �hen a subciliary approach �as perfor�ed. This indicates lo�er lid retraction, �hich did not see� to occur in a significant manner, when a transconjunctival approach �as perfor�ed (see Table 2). �n this study one ectropion �as observed. This �ay be related to the lo�er nu�ber of patients included in this study under�oin� a subciliary approach. �n previous studies si�ilar or even lo�er rates of ectropion �ere observed. Overall these results are endorsed by the present literature: Lo�er eyelid retraction is the �ost co��on co�plication after a subciliary approach (20,21). Scar contracture, cicatricial connection bet�een the septu� orbitale, orbicularis �uscle and surroundin� tissue as �ell as loss of �uscle tonus �ay provo�e scleral sho� and ectropion. Thus �ost authors prefer the transconjunctival approach (4,6,1�,20,22-24). Transconjunctival approaches reduce co�plications such as ectropion to a �ini�u� (2), but include the hi�hest ris� of entropion (�). Durin� the past decades the transconjunctival approach sho�ed an uninterrupted increasin� use. Alto�ether transconjunctival incisions see� to include a lo�er ris� of postoperative lo�er lid retraction and ectropion co�pared to transcutaneous and especially subciliary approaches, as su��est our data (see Table 2). The classification of orbital floor fractures investigated here did not yield significant influence on the eyelid �orpholo�y in our study. Previous analyses investi�ating other classifications of orbital floor fracture localizations reconfirm this result (4). Alto�ether this �ay be interpreted as evidence, that a postoperative lo�er eyelid �alposition is �ore dependent on the selection of the sur�ical approach than on the localization and type of the fracture. �n our center �e prefer the transconjunctival approach �henever possible. To our e�perience, the rate of ecor entropion is related to ine�perience. The level of the incision in the forni� is enor�ously relevant. The preservation of the septal inte�rity as provided by the retroseptal incision see�s �ost li�ely to us to prevent lo�er eyelid distortion (1�). We do not see indications for a transcutaneous approach in isolated blo�-out fractures, �hich are all satisfactorily accessable throu�h a transconjunctival approach. Only in case of �ore-fra��ent-fractures of the inferior or lateroinferior orbital ri� re�uirin� e�tensive e�posure �e do see indications for a transcutaneous approach in the for� of a subtarsal approach. The incision of the subtarsal approach should be placed as close as possible to the inferior border of the tarsal plate.The subtarsal approach �as jud�ed to be cos�etically acceptable �hen concealed �ithin a rhytid and less ris�y in �atters of lid retraction than subciliary approaches (20-22,2�-28).

Conclusion
Analyses of orbital fractures repair results should clearly distinguish isolated and combined orbital floor fractures. The evaluation of the effects of isolated blo�-out fractures and their operative therapy on the periorbital architecture by usin� anthropo�etric data e�tracted fro� standardized photo�raphs is reliable and ade�uate.
The subciliary approach exhibited a significantly higher rate of lo�er lid retraction than the transconjunctival approach.