Watering eye is a common complaint and may affect patients of any age. The term epiphora refers to the overflow of tears on to the cheek.
The complaint by a patient of a 'watering' or 'watery' eye or 'tearing' may not imply that tears actually overflow from the eye.
It is important to establish the underlying cause through a detailed history and appropriate clinical examination.
Epiphora does not automatically signify a lacrimal drainage system obstruction.
Epiphora is disabling because it interferes with the activities of daily living and may cause social embarrassment.
The skin of the eyelid and cheek may become excoriated and sore. The eyelid skin may contract and cause a secondary lower eyelid ectropion.
Repeated wiping of the tears can result in laxity of the canthal tendons, further contributing to a lower lid ectropion.
Epiphora is due to hypersecretion of tears, lacrimal pump dysfunction or lacrimal outflow obstruction. Outflow obstruction is commonly at the level of the nasolacrimal duct through which tears pass from the lacrimal sac to the nose, beneath the inferior turbinate.
An acquired obstruction is usually an idiopathic condition where fibrous narrowing occurs, with a female-to-male ratio of 4:1 and an increased prevalence with age. There is a wide variety of infrequently encountered secondary causes including inflammatory conditions and tumours, which may be intrinsic or extrinsic to the lacrimal apparatus.
Congenital nasolacrimal duct obstruction is common in neonates with a prevalence of 6-20 per cent. Ninety per cent resolve spontaneously in the first two years and a significant proportion of the remainder respond to probing of the nasolacrimal duct and/or temporary silicone stent insertion.
Dacryocystorhinostomy or DCR aims to create an anastomosis between the lacrimal sac and the nasal mucosa, above the level of a nasolacrimal duct obstruction. A bony ostium must be created through the floor of the lacrimal fossa. External DCR became popular because it consistently reached success rates of over 90 per cent.
Recent developments in endonasal surgical equipment, particularly the rigid nasal endo-scope, and perceived disadvantages of the external approach to DCR, led to refinement of the endonasal approach.
External DCR leaves a cutaneous scar on the side of the nose, which may be a concern to patients. The dissection must be continued through the medial canthal structures to access the lacrimal sac, which may interfere with the blink mechanism important for tear drainage. The patient may also suffer a medial canthal dystopia if the medial canthal tendon is not repositioned correctly. Patients may struggle to wear glasses comfortably for a short time after surgery.
The endonasal approach avoids these problems and surgery is confined to the structures relevant to creating the new ostium.
Endonasal DCR is performed under general anaesthesia or under local anaesthetic with IV sedation. A nasal mucosal flap is raised to expose the bone of the lacrimal fossa floor, which is then removed using ronguers and a diamond burr.
Once the lacrimal sac is exposed it is opened in a vertical fashion to form anterior and posterior flaps which are laid open against the lateral wall of the nose. A silicone stent is placed via the lacrimal canaliculi to maintain the new fistula while healing takes place. Postoperatively, patients are given antibiotic eye drops for seven days, and nasal douching with a saline preparation is recommended for two weeks to clear crusting and clots from the surgical site. A topical corticosteroid and decongestant are used intranasally for five days. The silicone stent is easily removed in the clinic after six weeks.
The technique is suitable for all ages, although in young children the technique is more technically demanding due to smaller nasal passages.
There may also be compliance problems in postoperative care, which is important to the success of the procedure.
Laser-assisted endonasal DCR is available but success rates have been considerably lower than for the non-laser assisted surgical technique.
It is more suited to patients in whom anticoagulants cannot be stopped or in patients with bleeding diatheses.
The evaluation of patients with epiphora includes nasal endoscopy to identify anatomical anomalies and intranasal pathology such as nasal polyps.
An endonasal DCR permits the correction of many of these problems intraoperatively and the location of the bony ostium may be tailored to the local anatomy.
Success rates for primary endonasal DCR have exceeded 90 per cent. Endonasal DCR has seen a steady improvement in success rates since the first published series. Our own results over the course of the past five years are certainly as good as those achieved for external DCR.
One disadvantage of endonasal DCR is the occasional need for a nasal septoplasty to be performed due to a deviation of the nasal septum.
Endonasal instruments are fine enough to permit access to the surgical site in most patients, however, a small proportion have a significant nasal septal deviation that obscures the surgical site. A septoplasty may be performed at the time of surgery. This increases the risk of complications such as nasal septal perforation or collapse of the bridge of the nose, but these are very rare in the hands of the experienced surgeon.
The need for a septoplasty should be identified preoperatively and the patient counselled appropriately. The identification of septal deviation and its correction may also be seen as an advantage of endonasal DCR, as mucosal adhesion to the nasal septum is a cause of DCR failure.
Another disadvantage of the technique has been the limited visualisation of the lacrimal sac where abnormal pathology may not be visualised or a biopsy could not be taken. However, this has been overcome as the technique has evolved.
A failed primary procedure may be due to local anatomical factors, closure of the epithelial-lined fistula by fibrous tissue, adhesions to the nasal septum or middle turbinate and granuloma formation. The endonasal technique gives excellent access without the need for an external procedure and offers equivalent chances of success as a secondary procedure.
Endonasal DCR is an evolving technique with many theoretical and practical advantages over external DCR for appropriately selected patients.
It promises to be a more refined and possibly more successful form of DCR for patients with nasolacrimal duct obstruction.
The time taken to perform an endonasal DCR is equivalent to the external approach. The cost privately through BUPA is £3,120 and the cost on the NHS is £1,082.
Endoscopic DCR is performed by a range of ophthalmic surgeons who are expert in oculoplastic surgery and some ENT surgeons in the UK.
Mr Leatherbarrow is consultant ophthalmic and oculoplastic surgeon, Manchester Royal Eye Hospital and BUPA Hospital Manchester, and Mr McKeag is fellow in oculoplastic and orbital surgery, Manchester Royal Eye Hospital
Watkins L M, Janfaza P, Rubin P A. The evolution of endonasal dacryocystorhinostomy. Surv Ophthalmol 2003; 48: 73-84.
Goldberg R A. Endonasal dacryocystorhinostomy: is it really less successful? Arch Ophthalmol 2004; 122: 108-10.
Simon G J B, Joseph J, Lee S, et al. External versus endoscopic dacryocystorhinostomy for acquired nasolacrimal duct obstruction in a tertiary referral center. Ophthalmology 2005; 112: 1,463-8.
McDonogh M, Meiring J H. Endoscopic transnasal dacryocystorhinostomy. J Laryngol Otol 1989; 103: 585-7.
Becker B B. Tricompartment model of the lacrimal pump mechanism. Ophthalmology 1992; 99: 1,139-45.
Tsirbas A, Davis G, Wormald P J. Mechanical endonasal dacryocystorhinostomy versus external dacryocystorhinostomy. Ophthal Plast Reconstr Surg 2004; 20: 50-6.
Malhotra R, Wright M, Olver J M. A consideration of the time taken to do dacryo-cystorhinostomy (DCR) surgery. Eye 2003; 17: 691-6.
Tsirbas A, Wormald P J. Mechanical endonasal dacryocystorhinostomy with mucosal flaps. Br J Ophthalmol 2003; 87: 43-7.
Howden J, McCluskey P, O'Sullivan G, Ghabrial R. Assisted local anaesthesia for endoscopic dacryocystorhinostomy. Clin Experiment Ophthalmol 2007; 35: 256-61.