Ocular Neostigmine Drops for Diagnosing Myasthenia Gravis

1Division of Pediatric Neurology, College of Medicine, King Saud University, Riyadh, Saudi Arabia 2Sinnar Hospital and Kush Eye Center, Omdurman, Sudan 3Basic Medical Sciences Department, College of Medicine, King Fahad Medical City, Riyadh, Saudi Arabia 4Department of Ophthalmology, College of Medicine, King Saud University, Riyadh, Saudi Arabia 5Department of Zoology, College of Science, King Saud University, Riyadh, Saudi Arabia 6Department of Medicine, Faculty of Medicine, University of Khartoum, Sudan 7Department of Pediatrics and Child Health, Faculty of Medicine, University of Khartoum, Sudan 8Department of Surgery, College of Medicine, King Saud University, Riyadh, Saudi Arabia


Introduction
Myasthenia gravis (MG) is an acquired autoimmune disorder caused by autoantibodies directed against epitopes on or around the acetylcholine receptor (AChR) in the muscle membrane [1]. The condition is characterized clinically by variable ptosis, ocular motility abnormalities, and weakness of voluntary muscles in other parts of the body [1]. Nearly all patients present with extraocular muscles involvement, causing ptosis and diplopia, and in 20-40% of patients, weakness remains limited to these muscles (ocular MG) [1][2][3]. The diagnosis may be difficult to make on clinical grounds especially in juvenile cases when weakness is restricted to few muscles such as in ocular MG [4], and a variety of supportive diagnostic tests have been devised [5].
Perhaps the best known of these is the Tensilon test [5][6][7][8], using intravenous edrophonium chloride, which is an acetylcholinesterase (AChE) inhibitor with very short action. Unfortunately, edrophonium causes cholinergic side effects that may include increased oral and bronchial secretions, bradycardia, abdominal cramps, sweating, heart block, hypotension, and red, watery, painful eyes. These are often uncomfortable for the patient, and although serious complications of bradycardia and syncope are rare [9] the test is now commonly performed with cardiac monitoring and atropine available as an anticholinergic agent. Other relative contraindications for performance of edrophonium are cardiac dysrhythmias and bronchial asthma. A second quick bedside test is by applying an ice pack locally to a ptotic eyelid for 2-5 minutes and observing for any improvement of ptosis or eye movement deficit [10][11][12]. The ice pack test is simple, inexpensive, and has a reported sensitivity of 93-97% and specificity of 97-98% [13]. This adjunctive diagnostic test is particularly helpful if the edrophonium test is contraindicated or not available. However, the procedure is unlikely to be tolerated by young children [14].
Other supportive tests require laboratory and electrophysiologic facilities. Serologic tests include assessing AChR antibodies which are found in 80% of adults with MG [15], 70% of peripubertal juvenile MG but only 50% of prepubertal MG [16,17]. Antibodies to muscle-specific kinase (MuSK) are rare in juvenile MG [18].

Electrodiagnosis of MG includes repetitive nerve stimulation [RNS]
(looking for a decrement of greater than 10% of the fifth compared to the first-evoked compound muscle action potential [CMAP] in multiple muscles). However, the test is uncomfortable in children, is artifact prone and can be abnormal in disorders other than MG [14]. The reported sensitivity of RNS ranges between 53% and 100%, but it is frequently negative in ocular MG [3]. Single-fiber electromyography (SFEMG) requires a significant degree of cooperation and expertise [18] and is not available in many hospitals. In young children, sedation may be required to facilitate both RNS and stimulated SFEMG [19]. An additional simple, safe, sensitive, and specific diagnostic test for MG is, therefore, needed.

Methods
This research project investigated the safety and efficacy of using intravenous (IV) neostigmine solution, available in all hospitals where anesthesia is performed, as an ocular drop for diagnostic purposes in the setting of patients with probable or definite autoimmune MG.
The study had two components: 1) experiments using 30 rabbits in which the standard IV neostigmine solution was instilled into one eye (using the other as control) to evaluate effectiveness (i.e., changes in the pupillary size of the experimental eye) and safety (i.e., does not cause significant irritation of the experimental eye or obvious systemic effects). 2) experiments in which the standard IV neostigmine solution was instilled into both eyes of patients with probable or definite symptomatic MG.
Both sets of experiments used the standard neostigmine (2.5 mg/ml) solution available in hospitals for anesthetic/surgical use for reversal of neuromuscular block. This solution is buffered and FDAapproved for use in humans. In both sets of experiments, one drop of this solution was used in an eye, limiting substantially the amount of drug that might be absorbed systemically.

Rabbit studies
Male New Zealand White rabbits (2.5-3.0 kg body weight) were used in all experiments. The animals were kept in a room of controlled room temperature (24.0+1.0°C) and humidity and a fixed dark light cycle (light on at 7:00 am and off at 19:00 pm).
The sterile neostigmine solution (2.5 mg/ml) was instilled into the right eye (one drop), while the control eye (left eye) received one drop of sterile saline (0.9% sodium chloride). Assessment of pupillary size was done at 30 minute intervals following instillation of neostigmine. The left eye of each rabbit acted as control for the right one.
The amount of neostigmine methylsulphate was calculated to be 0.1 mg/drop. Therefore, each rabbit's eye was instilled by 0.1 mg of sterile neostigmine methylsulphate. The sterile dropper for instilling neostigmine was used only once and then discarded. To prevent the drop of neostigmine from leaking out of the rabbit's eye, the lower eye lid of the rabbits was held out for at least 2-3 minutes in order to keep the drop inside the eye and allow for better absorption of the drug through the cornea.
The pupil size of both the right and left eyes of each rabbit was measured at baseline in mms by using a plastic ruler. After instilling the neostigmine solution (right eye) or saline (left eye), the pupil size was measured again at 30 minute intervals following drug or saline instillation. One drop of neostimine was instilled daily for 7 days and signs of ocular inflammation were observed.

Statistical analysis:
The mean values for the pupil sizes of the right eyes were recorded as well as those of the left eyes, separately. Paired t-test was used for comparison of data among the different groups on SPSS software. A probability value of <0.05 was considered to be significant.

Human studies
Patients with probable or definite MG were asked to sign informed consent approved by the appropriate Institutional Ethics Committees. Ptosis was recorded photographically, and pulse and blood pressure were also recorded prior to beginning the test. One drop of standard IV neostigmine solution (2.5 mg/ml) was instilled into each symptomatic eye of the patient. Fifteen minutes later, possible changes in ptosis were assessed. Pulse and blood pressure was recorded again and after another 15 minutes. At 30 minutes, the patient was evaluated also for possible changes in ptosis (documented by photography), obvious irritation of either eye and for report of discomfort.

Rabbit studies
At baseline (Figure 1) there was no significant difference in the mean [SD] pupillary size of the rabbits between the RE (7 mm [1.07]) and LE (7 mm     A representative photograph of these pupillary changes is shown in Figure 2.

Patients
Six patients (3 males and 3 females) who presented with ptosis and/or ocular motility disturbance were enrolled in the study (Table  1). Their ages ranged between 4.5 and 55 years (median=42 years, mean=36.4 years). The duration of their symptoms ranged between 1 and 18 months (median=4 months, mean=5.8 months).Three of them were not suspected before or investigated for MG, and hence received no treatment (Patients 1, 2 and 5). The other 3 (Patients 3, 4 and 6) were seropositive for AChR-antibodies, and one (Patient 6) had, as well, significant decrement (>10%) on repetitive nerve stimulation (RNS). Patient 3 ( Table 1) had prednisone and pyridostigmine for 3 months, another (Patients 4) had prednisone for 2 months, and a third (Patient 6) had pyridostigmine for one year. Patient 3 received 30 mg of pyridostigmine 2 hours prior to the test whereas Patient 6 had 60 mg of pyridostigmine 8 hours earlier. The 6 patients manifested clinically with ocular MG, with no limb, axial muscles, oropharyngeal or respiratory muscle involvement [5].
All 6 patients with MG had an observable increase of the palpebral fissure height (documented by photography) of at least 2 mm, 30 minutes after neostigmine instillation (Figure 3). The response was dramatic in the 3 patients with no prior treatment for MG (Patients 1, 2 and 5).

Discussion
Population based epidemiological studies in MG reported an incidence rate ranging from 1.7 to 21.3 cases per million person per year and prevalence rate ranging from 15 to 179 per million [20]. In both sexes, the incidence rate was reported to increase with age, peaking between 60 and 80 years, being more predominant in the older age group [20]. The proportion of childhood-onset (<15 years) MG appears to be relatively high in Asians [21]. Also in a series from Saudi Arabia [22], children (≤ 16 years) accounted for 20% of 150 patients who had thymectomy for the treatment of MG. The incidence of juvenile patients who present with ocular MG has been reported to range from 27% to 93%, with substantially high incidence of ocular presentation in Asian children [18]. A trend towards developing ocular MG rather than generalized MG was reported from South Africa in children of African ancestory [23], with an age peak between 2 and 4 years [24]. It's noteworthy that under-recognition of MG is likely in Developing Countries where access to specialist services is poor [25].
Although the clinical course of MG is variable, spontaneous remission is still uncommon and the time of diagnosis from onset (<1 year) and intervention was reported to be the most robust prognostic factor for remission [26]. Nevertheless, clinical diagnosis may be difficult especially when weakness is restricted to few muscles such as ocular MG.
The two bedside tests with high sensitivity and specificity include the ice pack and edrophonium tests. Nevertheless, application of ice over the eyelid for 2-5 minutes is unlikely to be tolerated by children (personal observation). Edrophonium test, which requires IV infusion of the AChE inhibitor edrophonium chloride, has a reported sensitivity of 71.5-95% and specificity of 97% [2,27]. The sensitivity of the test is likely to be lower in the busy clinical practice [2]. Moreover, false-positive tests (which have been reported with Guillain-Barre syndrome, brain stem lesions and compressive cranial neuropathies) might be higher in these settings [2,27]. Due to the potential serious cholinergic effects, edrophonium test in children should be conducted in a monitored setting with availability of atropine and staff trained in cardiac resuscitation [1,18,19]. Some patients with congenital myasthenic syndromes may deteriorate when exposed to an AChE inhibitor such as COLQ and MuSK mutations [28,29]. Also edrophonium is not    readily available in many hospitals of the Developing Countries. This is what prompted one of the of Authors (Mo A S) during his work in a district hospital in Sudan, to pioneer the idea of ocular neostigmine as a diagnostic test for MG. The neostigmine preparation (2.5 mg/ml) which has been used as ocular drops in the present study is available as IV solution (2.5 mg/ml) in all hospitals where anesthesia is performed.
The rabbits were chosen to test for the efficacy and safety of neostigmine eye drops since they are established experimental model for research on MG [30,31]. The rabbit experiment was included to show that the concentration of the drug is in the correct range, and that the drug does not cause irritation. Significant miotic effect of ocular neostigmine could be demonstrated at 30 and 60 minutes following its instillation (Figures 1 and 2). This significant miotic effect was also evident at 90 minutes. No ocular inflammation was observed even after administration of neostigmine daily for 7 days.
The application of neostigmine drops for 3 patients, who received no prior treatment for MG because the disease has not been suspected before, showed dramatic response ( Figure 3). It has been reported that MG is likely to remain restricted to the ocular muscles if the patient had restricted ocular manifestation of the disease for more than 2 years, similar to what has been observed in Patient 5 (Table 1) who had ocular MG for duration of 3 years [2]. The other patients (Table  1), who were all seropositive for AChR antibody, had prior treatment for MG (prednisone and/or pyridostigmine) and the disease has also been confirmed by significant decrement on RNS in a 47-year-old female (Patient 6, Table 1) who had the longest duration of therapy (pyridostigmine for one year). Although the response to neostigmine drops was less dramatic than in the first 3 patients who received no treatment for MG, an observable increase of the palpebral fissure height of at least 2 mm was observed 30 minutes after neostigmine instillation ( Figure 3). This sub-group included a 4½-year-old girl in whom edrophonium test is expected to be met with technical difficulties, as well as safety reservations. No ocular or systemic side effects were observed in any of the 6 patients.

Conclusions
Based on the present experimental and clinical research, ocular neostigmine drops proved to be an alternative safe, simple and efficient diagnostic test for MG. It may fulfill the need in many regions around the world where the other standard tests for MG are either unavailable or technically not feasible.