In the present study, the efficacy of dexamethasone injections into the lateral rectus muscle for treatment of acute AIANP is described and compared with patients who did not receive these additional injections. The results show that all of the patients who chose to receive local corticoid therapy at their first visit recovered completely, and their recovery time was shorter than that of the patients who did not select for corticoid therapy (P < 0.001). To the best of our knowledge, treatment trials have not previously been conducted to evaluate interventions directed toward hastening or improving recovery of oculomotor cranial nerve paralysis.
In this study, vascular paralysis was listed as the research objective for several reasons. 1) The diagnosis of microvascular palsies is often presumptive and clinically based, since anatomical and laboratory evidence are not available. 2) The clinical presentation, disease course, and prognosis of vascular paralysis are similar to those of idiopathic ocular motor nerve paralysis [1–3, 5, 7, 16–18]. Richards et al. reported a spontaneous recovery rate of 71.6% (129/180) for vascular etiology and 64.5% (160/248) for undetermined etiologies [3]. Similarly, Park et al. reported a 60% successfully treated and improvement rate for idiopathic oculomotor palsy and 71% for microvascular palsies [7]. 3) The most common microangiopathy, diabetic peripheral neuropathy, has been shown to primarily affect peripheral nerves, and rarely involves motor nerves. Generally, at the more severe stage of peripheral neuropathy, progression to muscle paralysis is observed. However, most ischemic nerve palsy patients with diabetes do not report typical peripheral neuropathy [3, 16]. 4) In a retrospective case control study, use of aspirin, a drug that improves circulation, did not improve healing for ischemic nerve palsy. Therefore, based on the above reasons, we regard these vascular factors as risk factors of AIANP in the present study, rather than direct causes.
Idiopathic facial palsy, also known as Bell's palsy, is an acute, generally unilateral, paralysis or weakness of facial musculature that is consistent with peripheral facial nerve dysfunction. The etiology of Bell's palsy has been debated for many years. It has been proposed that viral infections, vascular ischemia, autoimmune inflammatory disorders, and heredity are underlying causes [8, 10]. Isolation of the herpes simplex virus type 1 genome from saliva and facial nerve endoneurial fluid, together with other research evidence, has supported a viral etiology that is widely accepted [8–11]. Many aspects of Bell's palsy are similar to those of idiopathic ocular motor nerve paralysis. For example, the acute or subacute onset, a relatively stable period of disease (about 2–3 months), paralysis symptoms are observed to gradually improve, recovery is typically observed within 3–6 months, and the spontaneous recovery rate is estimated to be greater than 70% [19]. Moreover, based on the scientific evidence regarding Bell’s palsy that supports the contributions of neural inflammation and secondary ischemia to a neural blockade of the facial nerve that leads to facial paralysis, corticosteroids have become a first-line treatment for acute therapy. For example, a cochrane review that compared corticosteroids with a placebo showed moderate- to high-quality evidence from randomized controlled trials that supported a beneficial effect from use of corticosteroids [12]. We hypothesize that investigations of the etiology and treatment methods for facial nerve palsy can serve as a reference for investigations of the etiology and possible treatment strategies of ocular motor nerve palsy. Correspondingly, the former investigations support the use of corticosteroids for treatment of AIANP.
In a retrospective study employing high-resolution MRI, Park et al. [5] observed that a subset of patients with sixth and third cranial nerve palsy exhibited corresponding responding nerve enhancement. They hypothesized that these data represent a manifestation of neuroinflammation. Among their cohort, nine patients were administered oral prednisone therapy with or without intravenous methylprednisolone. The latter was indicated for the oculomotor palsy patients experiencing headache symptoms. After treatment, eye movement returned to normal in nine of the patients, with headache symptoms also disappearing. Yang et al. [13] previously reported that 31/45 (71.1%) patients with acute oculomotor nerve palsy exhibited enhanced oculomotor nerves in MRI. Among the eight patients of the latter cohort that received 80mg of methylprednisolone injection therapy, their condition improved. However, the numbers of cases in these retrospective studies were small and controls were lacking, making it difficult to explain the effectiveness of corticoid therapy.
More recently, several authors have reported cases of isolated abducens paralysis in patients with COVID-19 [20–24]. A single case involving abducens paralysis without respiratory symptoms was also described [20]. Based on these data, it is proposed that COVID-19 is another possible cause of abducens nerve paralysis. The present study included four patients who experienced abducens nerve paralysis after COVID-19 infection. All four of the patients eventually recovered, yet the patients treated with dexamethasone injections exhibited a shorter recovery time.
In addition to risk factors such as hypertension, diabetes, and smoking, 52 (72.2%) patients in our study exhibited excessive fatigue phenomena prior to the onset of abducens nerve palsy. These phenomena included continuous long hours of work or watching a mobile computer, staying up late, insomnia, stress, etc. These phenomena especially affected younger patients. It is possible these conditions reduce a patient's immunity, which could represent an additional risk factor for abducens nerve palsy.
In the present study, conventional oral corticoid therapy was not used. This decision was made based on observations that the concentration of an oral corticoid at a lesion site is low, the effect of an oral corticoid is slow, and there are many side effects that can occur, especially among the elderly and children. In previous studies, oral administration of prednisone was not as effective as a local injection in treating ophthalmoplegia caused by acute stage thyroid-associated ophthalmopathy [15]. Similarly, in the treatment of ocular myasthenia gravis accompanied by severe extraocular muscle paralysis, the effect of oral corticoid therapy was often observed after one month, and the effect was poor[25]. In contrast, Shi et al. reported that the effect of a local injection of triamcinolone acetonide was observed within one week, and a better treatment effect was observed [14]. Therefore, in this study, local administration of corticoid was used to treat abducens nerve palsy.
It is generally believed that an injection into the extraocular muscles will provide a more direct route to the pathological site of muscle paralysis. We observed that a direct injection of dexamethasone into the lateral rectus muscle not only improved the successfully treated rate of acute AIANP, but also significantly shortened the course of disease. Richards et al. reported a mean spontaneous recovery time for AIANP to be 9.66 weeks, and only 36.6% of their patients recovered within 8 weeks [6]. In this study, the average time from the start of treatment to full recovery for the 39 patients in the study group was only 3.45 ± 1.49 weeks. Meanwhile, for 29 patients in the control group who completed follow-up for more than 6 months it was 12.1 ± 9.83 weeks. The significantly shortened course of disease reduces the impact of diplopia on the daily lives of patients, and also significantly alleviates patients' anxiety.
In the present study, the dexamethasone injections that were performed included a small amount of lidocaine. The latter was included to alleviate patient pain and to reduce patients’ fear of an extraocular muscle injection. There was concern expressed regarding a possible toxic effect of lidocaine on local muscles and nerves, or direct injury to the external ophthalmic muscles. In previous reports, some patients who used peribulbar or retrobulbar block anesthesia experienced paralytic strabismus[26, 27]. It was speculated that this may be related to myotoxicity of anesthetic drugs such as lidocaine, bupivacaine on muscles, or may be caused by direct injury to the extraocular muscles by insertion of a needle. However, there is no direct evidence to support these possibilities. Animal experiments have shown that the effects of local anesthetics on muscles are related to drug concentrations, and low concentrations of local anesthetics generally do not cause degeneration of extraocular muscles. Even with minor degeneration, it quickly recovers[28–30]. Moreover, in one of the largest retrospective studies conducted to date, 150 cases of postoperative diplopia following cataract surgery were mainly because decompensating pre-existing strabismus. Use of topical anesthesia did not abolish this surgical risk. Thus, diplopia after cataract surgery may not be caused by the toxic effects of anesthetic drugs or muscle damage caused by injections. Moreover, extraocular muscle injections of botulinum toxin have been widely used in the treatment of strabismus. So far, there have been no reports of injection induced extraocular muscle injury [31, 32].In the present study, the paralyzed lateral rectus muscle recovered after treatment and no aggravation of muscle paralysis was noted. However, further studies are needed to confirm that lidocaine does not mediate any toxic effects following its injection into extraocular muscles.
In a retrospective study spanning 2–15 years and including 59 patients with presumed microvascular sixth nerve palsy, 16 patients (31%) experienced recurrent events either in the same or contralateral eye [33]. In the present study, recurrence did not occur among the patients who had recovered during a mean follow-up period of 17.4 ± 8.48months (range: 6–36).
Considering ethical issues, patients' wishes, and the research conditions established, the current study does not represent a randomized double-blind control study. In addition, the number of cases examined is small. Therefore, a large sample multicenter randomized controlled study is needed to further clarify the effects and complications of local corticoid injections on AIANP.
In conclusion, the results of this study indicate that staying up late, fatigue, or mental stress are contributing factors to patients with acute AIANP. A direct injection of dexamethasone into the extraocular muscles did improve recovery and shorten the course of disease in our cohort. Thus, our results suggest that location and mode and dosage of local corticoid therapy warrant further study, as well as the potential effect of local corticoid therapy on other ocular motor nerves. At the same time, the results of this study provide valuable insights which stimulate further explorations of the cause of AIANP.