Purpose
To determine the efficacy of the Fasanella-Servat operation in patients with moderate to severe blepharoptosis.
Methods
Case-review of 40 eyelids with an upper eyelid to pupillary Margin Reflex Distance (MRD1)
Results
The mean pre-operative MRD1 was 0.02 mm while the mean post-operative MRD1 was 3.14mm. There was one case of eyelid contour abnormality and asymmetry. Three separate complications occurred. One revision was required.
Conclusion
The Fasanella-Servat operation may provide functional improvement and serve as an alternative to traditional surgical options for those with moderate to severe blepharoptosis.
Blepharoptosis; Fasanella-Servat operation; Ptosis
In 1961, Fasanella and Servat introduced a simplified method for the correction of small amounts of blepharoptosis associated with good levator function [1]. They described the procedure as an excision of tarsoconjunctiva, Müller’s muscle, and the levator palpebrae superioris muscle. The Fasanella–Servat operation has been employed for all types of blepharoptosis, including congenital, myogenic, neurogenic, traumatic, ptosis in Horner’s syndrome, and blepharoptosis of the an ophthalmic socket. Regardless of the type of blepharoptosis, it is generally accepted that this procedure should be reserved forces in which there is at least 10 mm of levator excursion and no more than 3 mm of ptosis present [1-5]. It may also be used in cases of contour abnormalities with little to no ptosis [2].
The procedure gained wide acceptance from the very beginning. However, the surgical technique, the proposed mechanism of action, and its indications have been challenged and modified on several occasions [2,6-8]. The Fasanella-Servat approach was initially regarded as a form of levator resection [9]. Several authors have since reported excellent success rates without resection of the levator aponeurosis or Müller’s muscle, positing that the effect of the Fasanella-Servat procedure is achieved by shortening of the vertical posterior lamella, wound contracture, and advancement of Müller’s muscle along the tarsus [4,6].
Conventionally, the Fasanella-Servat operation has been employed for minimal ptosis with good levator function, and this is reflected in the published literature regarding outcomes of the Fasanella-Servat procedure [1-5,7,10]. We were curious regarding the utility of this operation in patients with moderate to severe ptosis, and we hereby present our experience with the Fasanella-Servat surgical repair of blepharoptosis in these patients.
We retrospectively reviewed the charts of a single surgeon from June 2003 to December 2009 for all Fasanella-Servat operations performed. The surgical records were collected to determine patient gender, laterality of Fasanella-Servat procedure, blepharoptosis type, date of last follow-up, and availability of postoperative photographs for review.
Inclusion criteria and outcomes measured
Study inclusion criteria included an MRD1 value of <1mm, a levator excursion of >0mm, and no past history of ptosis surgery on the same eye. Those patients with neurogenic, myogenic, or mechanical ptosis were not included in this study.
Post-operative photographs were evaluated by a blinded ophthalmic oculofacial surgeon for eyelid height, contour, symmetry, and overall cosmesis. Before evaluation, the photograph of the patients’ eyes was adjusted such that the horizontal white to white distance was 10mm, then MRD1was calculated.
Outcomes were classified based on a previously published study [11]. Eyelid height was considered excellent if the MRD1 was ≥2.5 and ≤4 mm; satisfactory for MRD1 ≥1 and <2.5mm; and poor for MRD1 <1mm or >4mm. Symmetry was regarded as excellent for equivalent eyelid heights, satisfactory for a height difference of ≤1mm between eyelids, and poor for a difference of >1mm. Eyelid contour was judged on the arch of the upper eyelid margin: excellent results exhibited a flawless, smooth contour; satisfactory results showed minimal eyelid peaking, flattening, or lateral ptosis; and poor results demonstrated significant contour abnormalities. Overall cosmesis was subjectively graded on the patient’s complete postoperative eyelid appearance.
Technique
The upper eyelid was everted and approximately 0.5-1.0 cc of 1% lidocaine with 1:100,000 epinephrine was injected subconjunctivally above the superior border of the tarsal plate. The amount of tissue to be resected was based on a modified Putterman and Fett algorithm [12], where moderate ptosis underwent a tissue resection of approximately 8.5mm, and severe cases had +1.0mm additional resection for every 0.5mm change in lid height desired. Two curved hemostats were clamped horizontally tip to tip toward the distal border of the everted lid. Adouble-armed 6-0 nylon suture was passed proximal to the hemostat clamps in a repeating horizontal mattress fashion across the everted lid margin, passing through the tarsal plate, Müller’s muscle, and conjunctiva. The hemostats were then removed, and the excess tissue distal to the horizontal mattress sutures was excised by cutting along the line of tissue compressed by the hemostat clamps. The ends of the sutures on the conjunctival side were passed through the wound edges and out through the lid crease onto the skin. Each end of the suture was tied upon itself in close apposition to the skin.
In our study, the Fasanella-Servat operation was performed by a single surgeon on 40 eyelids of 24 patients. There were 10 female and 14 male patients. Out of the 40 eyelids operated on, 22 were on the right side and 18 were on the left side.
The mean pre-operative MRD1 was 0.02mm and the mean post-operative MRD1 was 3.14mm (Table 1). Inpatients with severe ptosis of MRD1 ≤0 (n=11), the average pre-operative MRD1 was -1.3mm and post-operative MRD1 was +2.5mm (Table 1). The maximum value of pre-operative MRD1 measured in the patients was -3mm and the maximum post-operative MRD1 measured was +4.5mm. On average, approximately 9-10mm of tissue was resected.

Table 1: MRD1 values of patients before and after Fasanella-Servat procedure.
The table above shows the absolute pre-operative (blue) and post-operative (red) values in millimeters along the x-axis and the patient number along the y-axis.
Excellent procedure outcome as determined by the MRD1 value of ≥2.5 and ≤4mm was seen in 32 cases (80%). A satisfactory outcome with MRD1 of ≥1 and 4mm was seen in 5 cases (12.5%). (Table 2)Analyzing only patients with severe blepharoptosis, 6/11 cases (55%) had excellent surgical outcomes, 3/11 (27%) had satisfactory outcomes, and 2/11 (18%) had poor surgical outcomes (Table 3). Most of the eyelids operated upon had an acceptable outcome based on evaluation by an oculofacial surgeon for eyelid height, contour, symmetry and cosmesis (Figure 1). There was one case of eyelid contour abnormality in the form of lateral peaking, and this required revision. Additional complications included inclusion cysts in two patients and ectopic lashes in one patient.

Figure 1: Pre-and post-operative comparison of eyelid position and cosmesis.
2 patients who underwent Fasanella-Servat operation for correction of moderate to severe blepharoptosis. The patient on the left exhibited severe blepharoptosis of the right upper lid with a negative MRD1, and moderate ptosis of the left upper lid. The patient on the right displayed moderate blepharoptosis bilaterally.
Upper row: pre-operative eyelid position. Lower row: post-operative appearance.

Table 2: Pool data for surgical outcomes following Fasanella-Servat operation for moderate to severe ptosis.
Excellent outcome corresponded to MRD1 value of ≥2.5 and ≤4mm, and was seen in 32 cases (80%). Satisfactory outcome of MRD1 of ≥1 and <2.5mm was seen in 3 cases (7.5%). Poor outcome of MRD1 value either <1mm or >4mm was seen in 5 cases (12.5%).
Table 3: Surgical outcomes following Fasanella-Servat operation for patients with severe ptosis with MRD1 ≤0 mm.
6 of 11 cases (55%) had excellent surgical outcomes with MRD1 value of ≥2.5 and ≤4mm. 3 of 11 cases (27%) had satisfactory outcomes of MRD1 ≥1 and <2.5mm. 2 of 11 cases (18%) had poor surgical outcomes with MRD1 either <1mm or >4mm.