Remember me
ORIGINAL ARTICLE
Year : 2022 | Volume
: 12
| Issue : 4 | Page : 415-422
The associations between central serous chorioretinopathy and muscle relaxants: A case–control study
Manish Jain1, Sunir J Garg2, Mohammad Khan3, Varun Chaudhary4, Dena Zeraatkar5, Dhanya Kurian6, Sarath Lal6
1 Department of Ophthalmology, Veer Chandra Singh Garhwali, Government Institute of Medical Sciences and Research, Srinagar, Uttarakhand, India; Department of Ophthalmology, NMC Hospital, Al Ain, United Arab Emirates
2 Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania, USA
3 Department of Health Research Methods, Evidence and Impact, McMaster University, Ontario, Canada
4 Department of Surgery, Division of Ophthalmology, McMaster University, Hamilton, Ontario, Canada
5 Department of Health Research Methods, Evidence and Impact, McMaster University, Ontario, Canada; Department of Biomedical Informatics, Harvard Medical School, Boston, Massachusetts, USA
6 Department of Ophthalmology, NMC Hospital, Al Ain, United Arab Emirates
Correspondence Address:
Dr. Manish Jain
Department of Ophthalmology, Veer Chandra Singh Garhwali, Government Institute of Medical Sciences and Research, Srinagar - 246 174, Uttarakhand
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/2211-5056.361975
PURPOSE: To evaluate the role of muscle-relaxants as risk factors for the development of central serous chorioretinopathy (CSC) - the second most common retinopathy in our settings; despite multiple risk factors seen in our patients, 21% were initially labelled as idiopathic.
MATERIALS AND METHODS: Retrospective case-control study at a tertiary hospital in the United Arab Emirates, where we reviewed the medical records of 273 patients with CSC examined between 2010 and 2019 for use of muscle-relaxants including tolperisone/eperisone, carisoprodol and gabapentin/pregabalin within a year of onset/recurrence of the disease. Intake of drugs with known association with CSC (including corticosteroids/sympathomimetics) was also recorded. Two hundred eighty-six subjects with adverse events seen at the same institute during the same study period served as controls. Odds ratios, Chi-Square tests and multivariate logistic regression were carried out to determine any associations with the muscle-relaxants and other pharmacological confounders - corticosteroids/sympathomimetics.
RESULTS: Muscle relaxants may increase the risk of CSC as evident on multivariate regression analysis (OR: 2.55; confidence interval [CI]: 1.208-5.413); the significance was retained on removing the 6 subjects who had corticosteroids/sympathomimetics (OR: 2.30; CI: 1.073–4.939). Univariate analysis yielded an OR of 2.52 for muscle relaxants (CI: 1.2149–5.2276), 2.96 for eperisone/tolperisone (CI: 1.3531–6.5038), and 6.26 for eperisone as an individual agent (CI: 1.8146–21.6252).
CONCLUSION: We found muscle relaxants to be associated factors of CSC regardless of inclusion of corticosteroids/sympathomimetics (P < 0.05). Among individual classes of muscle relaxants in this study, only eperisone/tolperisone posed a significant risk (P < 0.05). The vascular smooth muscle relaxation could be the possible mechanism that affects the choroidal blood flow and indirectly predisposes to CSC.
Keywords: Carisoprodol, central serous chorioretinopathy, eperisone, gabapentin, muscle relaxants, pregabalin, tolperisone
Central serous chorioretinopathy (CSC) typically presents with a circumscribed serous elevation of the retina often causing distorted vision with decreased visual acuity. While most cases resolve spontaneously, others have a protracted or recurrent course leading to visual loss. Corticosteroid use, psychological stress, hypertension, psychotropic drug use, and type “A” personality traits have all been implicated in causing CSC.[1] Patients with corticosteroid-associated CSC may have changes in mineralocorticoid receptor function.[2] In others, the etiology appears to be choroidal dysfunction, particularly involving choroidal circulation which is under autonomic control.[3],[4],[5],[6],[7],[8] Autonomic dysfunction such as heart rate variability and sympathetic-parasympathetic imbalance have been reported in CSC.[7],[8]
Many drugs associated with CSC have a propensity to modulate the vascular tone or choroidal blood flow (ChBF) through diverse mechanisms that include actions on the autonomic nervous system, as well as on vascular smooth muscles.[4],[5],[6],[9],[10],[11],[12],[13],[14],[15],[16] In our settings, 21% of CSC patients were idiopathic. This prompted us to explore the potential role of other drugs or comorbid conditions in CSC.
In this article, we explore the potential role of different classes of muscle relaxants including tolperisone/eperisone (class I), carisoprodol (class II), and gabapentin/pregabalin (class III) as independent associated factors for CSC. We included corticosteroids/sympathomimetics in our analysis to mimic the real-life scenario, where concurrent or sequential intake of multiple drugs or risk factors exists, but withdrawal of corticosteroids alone does not lead to remission.[17],[18]
Materials and MethodsWe obtained the approval from the Institutional Review Board of NMC hospital in Al Ain, UAE for this case–control study (Approval date: May 27th, 2018). The study adhered to the tenets of the Declaration of Helsinki. The medical records of all CSC patients seen over a 10-year period (2010-2019) were explored for past medical history and medications. For our broader epidemiological study, we called the patients if they had missing information.
Participants
The clinical diagnosis of CSC was based on symptoms, such as decreased vision or visual distortion evident as central scotoma with or without metamorphopsia or micropsia. The diagnosis was confirmed by the presence of serous retinal detachment on fundus and optical coherence tomography examinations (3D OCT 2000 Topcon, Corp., Tokyo, Japan, or Spectralis HRA + OCT; Heidelberg Engineering, Heidelberg, Germany). Wherever deemed necessary, we obtained fundus fluorescein angiography to demonstrate active angiographic leakage (TRC-50DX; Topcon Corp., Tokyo, Japan).
Of the 329 patients identified, those with incomplete records, lost to follow-up, or with alternative provisional diagnoses including age-related macular degeneration, diabetic retinopathy, optic disc edema, Vogt–Koyanagi–Harada syndrome, or posterior scleritis were excluded; the remaining 273 patients diagnosed with CSC were included for further analysis [Figure 1]. The presence of recent psychological stress, comorbid conditions, all medications taken within the year preceding the initial diagnosis, or recurrence of CSC was recorded. The period of 1 year was based on a Taiwanese study that looked at the demographic characteristics, comorbidities, and corticosteroid use within 1 year before CSC diagnosis.[19]
Figure 1: Flow diagram showing the progression to the final number of CSC patients and controls. CSC = Central serous chorioretinopathyControls
All patients (338) with any systemic adverse drug event (ADE) seen at our hospital and reported to the pharmacovigilance department during the same study period (2010–2019) were reviewed; those with CSC (total 17) were excluded. Subjects below 20 years (28 patients) or over 60 years (7 patients) were also excluded in order to pair the patients with age-matched controls [Figure 1]. The remaining 286 subjects served as controls. UAE has a mobile population with a skewed sex ratio and a high number of immigrants. However, the immigration policies for specific nationalities vary over a period. As controls, the ADE cases were subjected to the same immigration policies over the study period, as the CSC patients. Hence, we believe that both the groups were drawn from the same reference population.
For both the case and control groups, a medication log was generated that included muscle relaxants as well as drugs associated with CSC during the study period.
Drug interactions and clinical pharmacology
Potential drug–drug interactions were explored by direct literature search as well as Medscape (Medscape, New York) drug interaction checker.[20],[21] Keywords included muscle relaxants, corticosteroids/sympathomimetics, and other co-medications used by these patients [Table 1].[22] In addition to the mechanism(s) of action, we also looked at adverse events such as orthostatic hypotension and other clinical effects that could indicate actions on vascular smooth muscle as these could alter cerebral and choroidal circulation. The most common group of medications taken by our patients was the proton-pump inhibitors. Two patients (patient 19 with eperisone and patient 23 with tolperisone) and one control (patient 8) had taken tramadol, a centrally acting opioid analgesic. Neither proton-pump inhibitors nor tramadol had any interaction with eperisone/tolperisone on our search.[20]
Table 1: Muscle relaxants and other co-medicationsf intake by central serous chorioretinopathy patientsStatistical tests
A multivariate logistic regression model was used to examine any association between the occurrence of CSC and the intake of muscle relaxants and other pharmacological confounders – corticosteroids/sympathomimetics. Chi-square test was used to determine the significance of predictor variables and associated odds ratios (ORs) of estimates were computed alongside the 95% confidence interval. We repeated the analysis after removing the six subjects who had corticosteroid and/or sympathomimetic prior to the onset of CSC to see if muscle relaxants retained a significant association with CSC. The ORs were calculated for muscle relaxants as a group, and this was followed by the ORs for three classes and individual drugs.
ResultsThe mean age of the patients was 39.7 ± 7.3 years with age (range: 21 years–59 years). In comparison, the average age of a control was 38.5 ± 9.6 years. The difference between the two groups was not significant (P = 0.11). All patients on muscle relaxants were male. All except one were either outdoor workers or had an occupational predisposition for prolonged standing/other strenuous postures such as squatting during fieldwork. One person was on muscle relaxants following a sports injury.
Twenty-five of 273 patients and 11 of 286 controls in the ADE group had muscle relaxants. The timeline of muscle relaxant use (plus corticosteroid/sympathomimetic) in the CSC and the control group ADE is presented in [Table 1] and [Table 2].
Table 2: Adverse drug event cases (controls) and their intake of muscle relaxants at the time of event or within a yearNineteen patients marked as “A” in [Table 1] either had no corticosteroids/sympathomimetics, or these were taken only after the diagnosis of CSC (patients 12, 13, 15, and 16), ruling out a primary causative role. Three patients [”B” in [Table 1]] had at least one of these two classes of drugs before consuming muscle relaxants, while in three additional cases [”C” in [Table 1]], the exact timeline of corticosteroids/sympathomimetics could not be established as prescriptions originated elsewhere and only the medical history or pills were available. The six cases marked as “B” or “C” were considered confounders for the multivariate analysis.
Among the controls, we had eight cases that had ADE related to muscle relaxants. Only three subjects had an ocular or visual adverse event attributed to muscle relaxants: controls 1 and 3 had allergic manifestations that involved conjunctiva and periorbital edema; control 7 on pregabalin had visual disturbance, but his ocular examination was noncontributory. While looking up at medications within a year, we picked up three additional controls who had muscle relaxants.
The ORs for muscle relaxants, their classes, and individual drugs are summarized in [Table 3]. As shown, the OR was highest for eperisone (6.26; P = 0.003). The OR for muscle relaxants was 2.52 (95% confidence interval [CI]: 1.2149–5.2276; P = 0.01). On multivariate analysis, the OR for muscle relaxants was 2.55 (95% CI: 1.208–5.413; P = 0.01). On removing the six confounders, the OR (2.30) was still significant (95% CI: 1.073–4.939; P = 0.02). [Table 4] summarizes the results of the regression analysis with and without the confounding drugs.
Table 3: Odds ratios for individual muscle relaxants (univariate analysis) Discussion CSC can cause mild-to-moderate vision loss in younger patients, and its etiology is not well understood. While corticosteroid use, stress, and a type A personality have been implicated in some patients, in many others, there is no identifiable cause. The current study suggests that for some patients, muscle relaxants may be a predisposing factor.
All 25 subjects who used muscle relaxants were male, which reflects the skewed gender ratio in the UAE due to a large number of immigrants, specifically men who perform manual labor.[7] Some occupations had predilection for male recruitment and their outdoor work may have increased the use of nasal corticosteroids/sympathomimetics in allergic rhinitis or of muscle relaxants due to strenuous activities or painful musculoskeletal conditions.
The three classes of muscle relaxants operate either through the voltage-gated channels (eperisone and tolperisone), or GABAergic pathways (carisoprodol, gabapentin, and pregabalin). Tolperisone and eperisone are centrally acting muscle relaxants that act at the reticular formation in the brain stem by blocking voltage-gated sodium and calcium channels. They inhibit spinal reflexes predominantly by a presynaptic inhibition of neurotransmitter release.[23] These drugs relax both the skeletal muscles and vascular smooth muscle. They affect vascular smooth muscles through blockade of adrenergic alpha-receptors.[21],[23],[24] Hypotension is a known adverse effect. They do not exhibit somnolence on withdrawal unlike carisoprodol which has an active metabolite meprobamate. Carisoprodol is a GABAergic central nervous system depressant that acts as a sedative and skeletal muscle relaxant.[25] It interrupts the neuronal communication within the reticular formation and spinal cord. Reflex tachycardia is a known adverse effect. Gabapentin and pregabalin have an affinity for alpha2-delta protein, an auxiliary subunit of voltage-gated calcium channels, and it modulates GABA and glutamate synthesis.[26]
Tolperisone and eperisone cause dilatation of basilar artery in guinea pigs.[27] Eperisone reverts the vasoconstrictive actions of norepinephrine, serotonin, and acetylcholine.[28] Tsokolas et al. described a case of vitreous hemorrhage ascribed to tolperisone.[29] Carisoprodol, gabapentin, and pregabalin have diverse cardiorespiratory, vasodilatory, and visual effects mediated through GABAergic pathways.[29],[30],[31],[32] Doğan et al. have recently reported two cases of CSC with pregabalin as a probable cause.[33] In vitro studies involving the isolated basilar artery of rabbit show vascular smooth muscle relaxation in response to GABA.[32] Gabapentin causes diverse ADEs such as somnolence, dizziness, headache, nausea, blurred vision, diplopia, altered color vision, macular edema, serous detachment, reversible visual field constrictions, and electrophysiological alterations.[33],[34],[35],[36] Another GABA analog, γ-vinyl GABA (vigabatrin), reduces the pulsatile ocular blood and pulse amplitude when used for epilepsy.[32] Pregabalin and gabapentin are structurally and functionally similar and reduce the synaptic release of many neurotransmitters.[26] A recent meta-analysis reports diplopia, blurred vision, and amblyopia with these drugs.[37] Previously, a role of serotonin, dopamine, and melatonin has been demonstrated in CSC,[12],[38],[39] while GABA has a role in the circadian rhythm and obstructive sleep apnea.[40]
Regulation of ChBF is complex and extends beyond the adrenergic pathways.[41],[42],[43] The sympathetic system employs adrenergic neurotransmitters and neuropeptide P, while the parasympathetic system uses vasoactive intestinal polypeptide, acetylcholine, and neuronal nitric oxide synthase.[6] Some of these affect the release of other neurotransmitters.[26] The presence of intrinsic choroidal neurons and other contractile cells provides additional mechanisms of ChBF control.[6],[44] The exact mechanism through which muscle relaxants may alter the ChBF remains conjectural but is likely to be mediated through their actions on vascular smooth muscles. Occasionally, drugs that do not cause significant hypotension per se may cause profound hypotension following drug interactions.[45]
ConclusionIn conclusion, our data suggest that muscle relaxants, especially eperisone and tolperisone, may be associated with predisposing factors for CSC regardless of consumption of drugs such as corticosteroids and sympathomimetics. There is a caveat: CSC has a known association with type A personality. People with type A personality are prone to stress, muscular tension and myofascial pain, and then need for muscle relaxant use. Compared to other potential controls, our choice of ADE subjects as controls may have some disadvantages: some of these might have had other systemic comorbidities that could have confounded the results. We do not have data on how well the groups were matched, especially with regard to ethnic origin, nor do we have specific data regarding other health concerns, presence of stress, or personality type. There were no female CSC patients that met the inclusion criteria. Additionally, the prevalence of muscle relaxant use is unknown, and often due to logistics and financial reasons, fluorescein angiography and repeat OCTs were not regularly obtained. However, this was the most complete set of accessible data available for controls at our institution.
The strength of the study is the inclusion of morbid conditions, medications, their interactions, and occupational and physiological background. More importantly, our observations open the possibility that other vasoactive drugs could contribute to CSC and this should be asked about on medical history. Further studies are needed to validate the effect of muscle relaxants on ChBF in other geographical areas and ethnic groups. Imaging techniques such as laser Doppler flowmetry and enhanced depth imaging could also be used to delineate choroidal vascular changes in response to vasoactive medications.
Acknowledgement
The authors acknowledge the contribution of Dr. Jay Chhablani, University of Pittsburgh, UPMC Eye Center, PA, USA through his valuable comments.
Financial support and sponsorship
Nil.
Conflicts of interest
The authors declare that there are no conflicts of interest of this paper.
References
Comments (0)