Functional Transnasal Endoscopic Conjunctivorhinostomy

Abstract

The various surgical options for lacrimal tract reconstruction include canaliculodacryocystorhinostomy; fundal transposition; Conjunctivodacrocystorhinostomy with Jones tube; and in extreme cases conjunctivo-rhinostomy (CR) for total lacrimal bypass. The first three require partial lacrimal tract presence/ integrity while CR is indicated when the canaliculi and/or sac are nonfunctional. This report describes a novel technique of minimally invasive CR through a predominantly endoscopic endonasal approach and further discusses its advantages over other established techniques.

Introduction

Injuries of the nasolacrimal complex are not uncommon referrals in clinical practice of otolaryngology particularly for epiphora. Trauma is reported to be the most common cause for the same particularly the road traffic accidents [1, 2] where the force impacts the upper part of bridge of the nose. Endoscopic dacrocystorhinostomy (DCR) is the most commonly performed procedure [3, 4] in cases of obliteration of lower lacrimal tract while for the upper lacrimal tract involvement, silicon intubation [5], canaliculo-DCR [6], fundal transposition or conjunctivo-DCR including synthetic tubes [7, 8] have been tried. A closed injury often masks underlying compound fracture with associated lacerations and avulsions of surrounding soft tissue [9]. In acute injuries with selective damage to lacrimal structures a microscopic reconstruction (such as canaliculo-DCR) may be possible, but in case of complete disruption, a neo-shunt directly between conjunctiva and nasal cavity (conjunctivorhinostomy) seems justified totally bypassing the nonfunctional lacrimal apparatus [10–12]. This report presents a much simple endoscopic technique with acceptable functional results and that too at a minimum cost/ technical-expertize and short hospitalization. Its comparison to the already established techniques is further discussed.

Case Report

A 25 year male with past history of head / facial trauma presented with marked epiphora and cicatrized wound overlying a healed compound fracture around left naso-lacrimal apparatus (CT scan Fig. 1). The vision/ eye movements were normal, but syringing revealed complete regurgitation with no discrete structure of lacrimal apparatus discernable (as appreciated on CT). Hence a CR was undertaken under general anaethesia. To define prospective site for neo-shunt, a meticulous blunt probing was done in conjunctival fornix to appreciate the weakest segment of lateral nasal wall corresponding to the thinnest bony partition (correlated with CT). Subsequently fibrotic tissue was excised around medial canthus and a thin ballpoint probe was negotiated towards nose in anteroinferior direction. Fortunately neo-osteogenesis was minimal while fractured segments were ‘segregated’ enough to leave a ‘mucosal window’ for probe to pass through fractured complex (arrow in Fig. 1). The point at which probe displaced lateral nasal wall and produced a bulge inside the nose was appreciated by nasal endoscopy and a suction cautery was used to vaporize the adjoining mucosa/ expose the underlying bone. Thereafter cutting burr was used to thin out the bony margins and remove the bone chips thus widening the path for probe to enter nasal cavity. The redundant mucosa as well as surrounding bone was further removed to further widen the passage. Thereafter while observing through nasal endoscope, a blunt (tip filed) wide bore needle (No. 16) carrying a 1/0 nylon suture was negotiated through this passage from medial canthus area into the nose. The other end of thread was delivered out of nose and the needle was subsequently withdrawn leaving a nylon thread in passage. For preparing a mucosal conduit, a 1*3 cm mucosa was incised from the inner aspect of lower lip (Fig. 2) and inversely sutured over an infant-feeding-tube segment using absorbable catgut suture (Fig. 3). The ends of so formed mucosal conduit were further sutured to the surface of the tube with catgut suture so as to avoid its sleeving during insertion. The (medial canthal) end of nylon suture was introduced through the infant feeding tube (carrying the mucosal graft) that in turn was negotiated along the passage with the nylon suture acting as a ‘guide-wire’. Nasal endoscopy confirmed the delivery of the mucosal graft inside the nose (Fig. 4). Care was taken for external end of tube not to project above medial canthus. Finally the 2 ends of nylon suture (first extruding through nose and second through medial canthus area) were loosely tied and hemostasis achieved. The patient was discharged the next day with an advice to continuously instill antibiotic eye drops and to attend regular follow ups. The in-vivo segment of infant feeding tube was spontaneously extruded through nose after 15 days and endoscopy confirmed the uptake of mucosal graft. The patient was advices to railroad the suture many times in a day for next 3 months after when it was removed (epiphora had resolved). The patient remained asymptomatic for next 6 months after which he was lost to follow up. However he did complained of minimal crusting in eye during waking only. No syringing was attempted through this conduit and no granulations could be seen.

Fig. 1.

Coronal CT scan depicting crimpled fractured bony segments with no element of nasolacrimal passage. The arrow indicates a thin bony partition overlying the mucosal window and the surrounding thickened bone

Fig. 2.

A rectangular mucosal graft obtaining from the lower lip

Fig. 3.

Mucosal graft tied over an infant feeding tube to form a conduit

Fig. 4.

Nasal endoscopy demonstrating the end of mucosal conduit (cross) wrapped over infant feeding tube

Discussion

Lacrimal surgery is continuously evolving with newer techniques focusing on better functional results. The conjunctivo-DCR with Jones tube, bypasses upper lacrimal tract with 83% success rate, but has been criticized [13] for the potential of granulation tissue formation, infection from incompletely apposed mucosal flaps, reduced flow of tears through scar tissue as a consequence of healing by secondary intention and a greater likelihood for bony regrowth across an ostium that is deficient in a tissue plane. Another known technique canaliculo-DCR is a more physiological reconstruction, but is limited by availability of sufficient canalicular structure and requires a lengthy/ tedious microsurgical dissection under microscope. Both the above approaches may not be possible in severe lacrimal scarring/ disruption and hence CR may be the only alternative. Apart from traumatic conditions, other indications of CR may include chronic dacryocystitis with blocked canaliculi and total or near-total lacrimal sac obliteration; or paresis of orbicularis oculi (nonfunctional pump mechanism) [6, 14–17]. Murube del Castillo [10] had initially described CR as a rapid office procedure under local anesthetic but showed high failure rates with permanent prosthetic stenting, prolonged postoperative maintenance and lengthy follow up. In order to avoid prolonged stenting and minimize infection, certain interesting modifications were suggested. Campell et al. [18] with a buccal mucosal graft found definite advantage over vein graft (in terms of obliteration), and found apparent reduction in occurrence of granulations/ mucosal overgrowth. However the disadvantages included a higher risk of failure with compromised recipient bed vascularity (due to irradiation, burn scarring); and difficult technicalities associated with medial canthal distortions (due to injury or prior surgery). Overall the literature reflects failure with free grafting of vein [19], as well as buccal mucosa [18] due to secondary intention healing, granuloma formation, and fibrosis resulting in restenosis possibly as a reaction to foreign stent being used simultaneously. Some alternative techniques using nasal and conjunctival flaps [20], pedicled nasal septal tube [10] and combination of conjunctival flap with cartilage from nasal septum [12] have also been described more recently. The preferred open CR has the advantage of bipedicled mucosal flap apposition (healing by primary intention), and temporary short-term stenting [13, 20]. While opthalmological assistance is required to mobilize conjunctival flap (avoiding corneal/ scleral injuries), an otolaryngologist is also needed for rotation of nasal mucosal flap, further making the surgery a team effort. A wider surgical defect is created to accomplish the results. The salient advantages and disadvantages of various surgical techniques are comparatively summarized in Table 1.

Table 1.

Comparison of surgical techniques for epiphora

Advantages	Disadvantages
Canaliculodacryocystorhinostomy
Physiological reconstruction	Limited by availability of sufficient canalicular structure Prolonged tedious dissection Need for microsurgical approach Not possible in severe lacrimal scarring/disruption
Conjunctivodacrocystorhinostomy with Jones tube
Bypasses upper lacrimal tract 83% success rate	Granulations Infection Scarring with reduced flow of tears Healing by secondary intention Chances of bony regrowth across ostium Not possible in severe lacrimal scarring/ disruption
Conjunctivo-rhinostomy: bypasses entire lacrimal system
Open conjunctivo-rhinostomy with vein graft
Donor site more manageable than buccal mucosal graft	More chances of obliteration as compared to free mucosal graft Mucosal overgrowth Healing by secondary intention Granuloma formation Fibrosis resulting in restenosis
Open Conjunctivo-rhinostomy with free mucosal graft
Reduced chances of obliteration than vein graft Reduced occurrence of mucosal overgrowth	Higher risk of failure with compromised recipient bed vascularity Difficult technicalities with medial canthal distortions Healing by secondary intention Granuloma formation Fibrosis resulting in restenosis
Open conjunctivo-rhinostomy with bipedicled nasal and conjunctival flaps
Mostly healing by primary intension Lesser chances of stenosis	Uncommonly healing by secondary intention and severe stenosis Secondary limitations in globe movements Chances of corneal and scleral injury albeit small Dislodgement and extension of tube Adhesions Granulation tissue formation
Open Conjunctivo-rhinostomy with pedicled nasal septal tube
Mostly healing by primary intension Lesser chances of stenosis	Uncommonly healing by secondary intention and severe stenosis Dislodgement and extension of tube Adhesions Granulation tissue formation Nasal and conjunctival irritation
Open conjunctivorhinostomy with combination of conjunctival flap with cartilage from nasal septum
Mostly healing by primary intension Offers autologous stenting and so least chances of restenosis	Dislodgement and extension of tube Adhesions Granulation tissue formation Nasal obstruction due to rotated nasal cartilage flap Nasal crusting may occur at narrowed site necessitating repeated nasal douches Second surgery for excision of nasal septal cartilage pedicle
Functional Endonasal Endoscopic Conjunctivo-rhinostomy with free mucosal graft
Minimally invasive No external scar Least size of nasal trephination as opposed to much larger osteal window in other techniques Minimal need of ophthalmologist Minimal chances of granulations & infection with nylon thread No chances of globe morbidity Minimal hospitalization Best cost effective analysis No special technical training needed	Chances of minimal stenosis if any Temporary cosmetic problem as patient needs to wear a nylon suture thread on face for 2 months Lip tenderness for 1 week at donor site of graft Nasal crusting along the nylon thread necessitating repeated nasal douches to prevent infection Repeated railroading the thread to clear off any retentions

Our technique is minimally invasive, most cost effective, requiring a maximum of overnight’s hospitalization and can be carried out solely in a rhinology setup (otolaryngologist trained in basic FESS) with practically no assistance from ophthalmologist. Being minimally invasive the chances of infection, scarring and granulations are also minimal; while being absolutely away from the bulbar conjunctiva there are no chances of globe traction or scleral injury. The only important consideration is to define appropriate passage with better vascularity for mucosal graft to have best chances of uptake. It is important to note that since fornix is usually free from tissue trauma, fistulization through fornix may have a better outcome (due to better vascularity) than through the lacus. Secondly selection of the appropriate site with minimal cicatrix and bone regrowth is critical since a better vascularity is likely to enhance the chances of graft uptake. Moreover for better capillary action and drainage through this ‘physiologic’ conduit, the fistulous tract needs to be oriented more vertically. It is also possible to drill through thick neo-osteogenic segment of nasal bone for such orientation. The failures of free mucosal grafts as reflected in literature were possibly a consequence of either open (more extensive exposure for creating large osteal perforation) procedures predisposing to infection/ vascular compromise the sizable (large) stent for prolonged period predisposing for foreign body reaction, ultimately translating in delayed healing/ granulations. The use of a thin nylon suture with continuous railroading as in our technique is highly unlikely to produce a foreign body reaction. However the minimal morning crusting as seen in our case may reflect a minimal degree of stenosis that possibly compromised tear drainage only during night hours particularly in absence of active eye movements (inefficient lacrimal pump mechanics). This could have been prevented by using a wider carrier tube. In addition however the retention of nylon thread for 3 months on the face of the patient may not be cosmetically acceptable to some patients initially but can indeed be overlooked considering the long term results. Theoretically a longer conduit may provide better valve-mechanism to prevent regurgitation but in turn may increase epiphora; while shorter conduit may lessen epiphora but enhance regurgitation. None of the free mucosal grafts have been tried through this approach till date but with comparable results and technical ease this technique may be adapted as the most conservative reconstruction technique for CR.

Funding

None.

Declaration

Conflict of interest

The authors declare that they have no conflict of interest.