Introduction

Botulinum toxin is known to be very effective in patients with essential blepharospasm or other focal dystonias [16]. The duration of efficacy is about 8–10 weeks and is stable in long-term treatment [8, 10].

Nevertheless, about 4% of more than 2,000 patients treated in our institution have shown an insufficient effect of botulinum toxin therapy. Pharmacologic therapy as an alternative is often disappointing and may show undesired side effects [4]. Other surgical procedures (e.g. excision of orbicularis muscle or resection of facial nerve branches) are mutilating and often unsatisfactory with regard to long-term results and also reveal a high rate of side effects.

Within the group of poor responders to botulinum toxin, some patients suffer from a special subtype of blepharospasm called “apraxia” of lid opening, also known as “dystonic lid-opening disorder of the levator-inhibiting type”. In these patients orbicularis spasms are absent, patients try to open their eyes by innervating their frontalis muscle or by using their finger tip, like other essential blepharospasm patients. This gave us the idea of performing a frontalis suspension operation of the kind normally used in certain types of ptosis. Good results in the first 28 patients were encouraging [14], and nowadays we perform this procedure in patients with apraxia of lid opening as well as in patients with essential blepharospasm who are poor responders to botulinum toxin treatment. The purpose of the present study was the evaluation of long-term results, complications and patients’ satisfaction.

Methods

Between 1992 and 2004, we performed frontalis suspension in 252 eyes of 132 patients with essential blepharospasm. The patients’ mean age was 64.0±9.4 years (range 36–84 years; Fig. 1); 44 patients were male, 88 female. Forty-five patients had the levator-inhibiting type (apraxia of lid opening). The indication for surgery was determined by the patient’s socioprofessional disability despite botulinum toxin injections.

Fig. 1
figure 1

Age distribution of patients at surgery

Full size image

One hundred twenty patients received bilateral surgery, The other 12 patients either had bilateral complaints of different severity, wished to have surgery only in one eye or were not able to undergo surgery of the fellow eye due to other diseases. The surgical technique was that first described by Carlson and Jampolsky [5], as modified by Roggenkämper and Nüßgens [13]. Surgery was performed under local anaesthesia in 120 patients, while 12 patients were operated upon under general anaesthesia (mostly bilateral). Silk sutures were employed in the first 14 eyes, and in all others we used Gore-Tex suture material. It was explained to all patients before surgery that further injections with botulinum toxin would be necessary, as lifting the upper lids by means of the frontalis muscle against the force of the (paralyzed) orbicularis muscle is then facilitated.

The duration of blepharospasm before surgery ranged from 1 to 27 years (mean 7.0±5.1 years). In all patients botulinum toxin treatment was administered before surgery. Seventeen patients were transferred from other departments or ophthalmologists for surgery and were never injected with botulinum toxin in our department; however, we obtained information about previous treatment results. The mean number of botulinum toxin treatments before surgery was 16.6±12.2 (range 4–77), the mean dose before surgery 21±8 units per eye (range 10–42.5 units). About 30% of botulinum toxin injections showed no effect, while in the others the effect was brief and incomplete. Botulinum toxin treatment proved to be insufficient even if increasing doses and varying injection sites were used. Thirty-one patients underwent botulinum type B (Neurobloc™) therapy before surgery [15]. Twenty-six of these patients had one to four injections with no effect; six reported side effects such as dry eye or dry mouth. The remaining five patients reported a better but not sufficient effect with botulinum type B than with type A. Before surgery, seven patients were actually injected with botulinum toxin type B in doses between 625 and 1375 units per eye.

The postoperative assessment included: postoperative complications, further surgery on the eyelids and botulinum injections after surgery (number of injections, effect, actual dose, botulinum type A/B). The patients’ subjective improvement after surgery was graded on a scale ranging from 0%, meaning no improvement, to 100%, meaning no complaints. Patients were asked, either on their follow-up visit or by telephone, to grade their subjective improvement on this percentage scale.

The improvement could also be observed by the physician in the case of patients attending for follow-up visits: voluntary lid opening, even continuously and repeatedly, was possible on demand by using the frontalis muscle. Patients were free to give additional comments.

Statistical analyses of the data were performed with t-test in the case of normally distributed values with Mann–Whitney U-test Spearman rank correlations in the case of non-normal distributions. Tests for normal distributions were performed using the Shapiro–Wilk test. P values are given if statistical significance was reached (p<0.05).

Results

Forty-two of the 132 patients who had undergone surgery attended our department between January and June 2005 for botulinum toxin injection and were assessed directly. Seventy-three patients were interviewed in a standardized way by telephone (including the 17 patients who had been transferred from other ophthalmologists). Seventeen patients had moved and could not been reached or had died. In these cases we used the available information of the patient file up to the last visit for evaluation (e.g. complications up to last visit, botulinum toxin injections). Follow-up duration ranged from 3 to 154 months (mean 59.6±50.0 months; Fig. 2), 60 patients were followed up for at least 5 years. At the time of the examination, patients had suffered from blepharospasm for a mean 11.4±6.4 years (range 1–33 years).

Fig. 2
figure 2

Follow-up in months

Full size image

In nine patients there were postoperative corneal problems with either postoperative erosion of the cornea or keratitis due to incomplete lid closure. In all cases treatment with ointments was successful. No serious corneal complications occurred, although slight overcorrection is desirable in the first few days after surgery for a satisfactory long-term result.

In one patient the suture worked loose and was tightened under local anaesthesia 2 days after initial surgery. In 18 eyes of 14 patients (7% of eyes) suture granulomas or extruded suture material were surgically removed. All but two additional operations were performed within the first 18 months after primary surgery. Thirteen patients had dermatochalasis surgery since their frontalis suspension (mostly bilateral), and 10 had undergone other types of lid surgeries (entropion surgery of the lower lid, removal of papillomas, etc.) without direct correlation to the frontalis suspension. Details can be found in Table 1.

Table 1 Comparison between patients with the levator inhibition subtype and other forms of essential blepharospasm concerning patient characteristics, postoperative complications and additional surgeries
Full size table

Long-term subjective improvement showed a median of 50%. The distribution of the subjective improvement in the 115 evaluated patients is shown in Fig. 3. Eighty-four patients (73%) reported an improvement after the surgery. Of the 31 patients with no subjective improvement, 21 were assessed by telephone. Of the 10 clinically examined patients with no subjective improvement, the objective effect of surgery was good in five patients, but they had a lack of understanding of the principle of the surgery and did not adequately use their frontalis muscle. The other five patients had undergone up to three further operations (suture tightening, suture removals), and the effect was insufficient, but they currently did not wish to have further surgery. There was no significant influence in outcome of surgery (subjective improvement) between the patients with levator-inhibition type and other types of essential blepharospasm (Mann–Whitney U-test).

Fig. 3
figure 3

Subjective improvement

Full size image

Subjective improvement was independent of the patients’ age at surgery, the duration of blepharospasm before surgery or the actual duration, the year of surgery, anaesthesia, suture material, number of injections or botulinum dose before surgery, the use of botulinum type B and the proportion of injections without effect before surgery (Spearman correlations). There was even no correlation between the need for further operations and subjective improvement (Spearman correlation), although there was a tendency towards lower subjective improvement if one or more further operations had been performed.

Of the 115 patients with current information about treatment (17 could not be contacted; see above), 110 received regular bilateral injections with botulinum toxin after surgery. Doses for botulinum toxin A ranged from 10 to 42.5 units per eye (mean 21±8), for botulinum toxin B from 500 to 2000 units per eye (mean 1080±530; eight patients). Of the five patients currently not receiving injections, four said they did not feel any additional benefit of the injections (two patients with no improvement, two with good improvement of 70/80%), and one patient was currently hospitalised due to other diseases and therefore could not come for treatment. All 12 patients with unilateral surgery had bilateral problems and bilateral botulinum toxin injections before and after surgery. Five patients had bilateral complaints of different severity and wished to have surgery only in the predominantly affected eye (subjective improvements of 10%, 20%, 30%, 50% and 60%). In two cases we had no up-to-date information as the patients could not be reached, but these patients were initially planned to have surgery also on the second eye. One patient had progressive dementia due to Parkinson’s disease and the treating physician advised against surgery, and one patient developed a coagulopathy and the risk of bleeding was assessed as too high for surgery. Two patients had insufficient effect of the local anaesthesia and wished no general anaesthesia for surgery on the other eye (subjective improvement 0% and 100%). One patient has been operated upon recently (after closing of the evaluation); he had delayed surgery on the second eye due to personal reasons.

The effect of surgery generally remained stable over the years. In all cases of decreasing effect (5% of eyes, 13 eyes of 9 patients), the sutures were tightened under local anaesthesia.

Thirteen patients had at some time tried ptosis crutches (ptosis spectacles). The use of this device was independent of patient age, type of blepharospasm or other factors. There was no significant correlation with the subjective improvement achieved by the surgery.

Of the 28 patients of the study published in 1997 [14], 24 could be reassessed. Twelve patients with initial subjective improvement (in 1997) of 30–100% reported exactly the same effect, while in three patients the effect was now 10% lower (initial effect between 50% and 80%). Nine patients now had effects of 0–10%. Eight of these nine patients had also initially experienced subjective improvement of only 0–20%. Of these eight patients, two had had a lack of understanding, three had undergone revisions before 1997 and one had experienced secondary wound healing. In one patient with initial effect of 60% the effect decreased to 0%. This patient was one of the first patients operated upon in 1992 using silk sutures. In 1999, 2000 and 2005 re-operations had been performed due to a loss of effect. Thus 15 of 16 patients with initially satisfactory effect remained stable.

Discussion

Frontalis suspension is a proven method for treating patients with ptosis with low or absent levator muscle function and/or ptosis with absent Bell’s phenomenon. We demonstrated frontalis suspension to be a valuable procedure in patients with essential blepharospasm unresponsive to botulinum toxin, who use the frontalis muscle or alternatively a finger to keep their eyes open. It is impossible to evaluate a patient’s treatment response or improvement rate only objectively, and therefore it is difficult to make an objective comparison of the relative success of different treatment options for blepharospasm [3]. A patient’s subjective improvement remains the best indicator of treatment efficacy [2]. Subjective improvement occurred in 73% of our patients who underwent surgery. An observable effect occurred in an even higher proportion, as some patients were not able to fully understand the concept of the surgery and did not use their frontalis muscle in an adequate way.

Before botulinum toxin became available, Puttermann and Urist published an account of five patients treated successfully with a frontalis sling [12]. In another study, 13 patients with apraxia of the eyelid were treated with frontalis suspension, 10 of them successfully [7]. Our own study on 28 patients in 1997 confirmed the preliminary results of 1993 [13, 14]. We were now able to re-evaluate 24 of these initial 28 patients. The effect remained stable over many years, and from the patient’s viewpoint the improved quality of life is a central consideration.

The patients’ acceptance of the operation is high. Frontalis suspension is a short procedure which can be performed under local anaesthesia. In contrast to the surgical alternatives (e.g. excision of the orbicularis muscle or resection of facial nerve branches) frontalis suspension is non-mutilating. It can be considered, if required, as a reversible procedure. Complementary botulinum toxin injections should be administered. In such cases one should spare the frontalis region (to facilitate sufficient force of the frontalis muscle).

Another surgical approach is orbicularis myectomy. It is often used in cases of functionally impaired patients with blepharospasm who have not tolerated or responded well to botulinum toxin therapy [11]. The myectomy procedure developed by Anderson before botulinum toxin was generally available involves the complete removal of the orbicularis, corrugator superciliaris and procerus muscles [1]. Most authors recommend first only upper myectomy [2, 11]. Complications of orbicularis myectomy of the eyelids are periorbital ecchymosis, hematoma, swelling and chronic lymphedema of the periorbital region (which may take 6 months or longer to resorb), numbness of the forehead (which is temporary, but may last a year or longer) and excessive weakness of eyelid closure leading to exposure keratopathy [2, 11]. Following treatment, about 50% of patients require treatment with botulinum toxin [6]. Postoperative botulinum toxin treatment is more effective and wears off less quickly, but because of scarring, injections are more painful. Some patients require a second operation to remove muscle that was missed at the first operation, but there is also the possibility of injecting low doses of botulinum toxin in the remaining muscle [11]. In a retrospective analysis of 330 patients who underwent orbicularis myectomy before 1988, 88% of patients noted an improvement; the acceptance rate was 82% [2]. Only 34% of these patients had been injected with botulinum toxin prior to surgery, so these patients were less selected than our patients, who had all failed botulinum toxin therapy.

Other therapeutic options for patients with essential blepharospasm with poor response to botulinum toxin include facial nerve avulsion—which has a high recurrence rate and untoward side effects and is therefore at best useful in selected patients [3, 11]— and doxorubicin chemomyectomy. Doxorubicin is an antineoplastic agent which is injected repeatedly over a period of 1–2 years. Side effects are local swelling, bruising and pain; the therapy may result in permanent loss of subcutaneous fat, contraction and scarring [17].

A number of papers have recommended frontalis suspension surgery in cases of levator-inhibition type and orbicularis myectomy in cases of regular essential blepharospasm with insufficient effect of botulinum toxin [2], but to our knowledge there is no study comparing the two indications. Grivet et al. compared the outcome of both types of surgery, but the indication for one or the other type of surgery was not clearly stated [9]. The functional disability fell significantly whatever the clinical form of essential blepharospasm or the surgical procedure, and there was no significant difference between the outcome of frontalis suspension and that of orbicularis myectomy 3 months after surgery. The authors reported that 15 of the 107 patients who first had orbicularis myectomy later had frontalis suspension, but no details about the indication were given and there was no evaluation of complications. In their opinion, the need for botulinum toxin re-injections does not reflect surgical failure, as patients with low functional disability may request injections, while others with high functional disability may not. Unlike other studies, we did not try systematically to wean the patients off botulinum toxin injections, but recommended injections to all patients, and we told the patients before surgery that they would probably have to continue with the injections.

In our study, frontalis suspension was equally effective in patients with eyelid apraxia and in patients with essential blepharospasm who respond poorly to botulinum toxin treatment. Therefore, we believe that this very effective and minimally invasive intervention should not be restricted to patients with the apractic type.

Conclusion

In this study we demonstrated good long-term results of frontalis suspension not only in patients with apractic forms of lid opening, but also in patients with other forms of essential blepharospasm non-responsive to botulinum toxin. Frontalis suspension can be considered as a minimally invasive but very effective (improvement in 73% of patients) and even reversible procedure in poor responders to botulinum toxin with good long-term effect and good acceptance by the patients. Additional treatment with botulinum toxin is required in most patients in order to increase the desirable imbalance between the frontalis and the orbicularis muscle.