Visual field changes in patients receiving antitubercular therapy: A prospective cross-sectional study from South India

Introduction

Tuberculosis (TB) is one of the ancient diseases known to mankind that has co-evolved with humans for many decades¹ and is a global health challenge. TB is one of the top three infectious killing diseases in the world and is also the leading cause of death from a single infectious agent.² India is a country with the world’s largest TB burden, accounting for one-fourth of global TB incidence.³ According to the India TB Report 2022, there is a 19% increase in cases (new and relapse) in 2021 from 2020.⁴ Interrupting its transmission is central to achieving the reduction in its incidence required to meet the End TB targets.⁴

DOT (Directly observed therapy) is widely accepted and followed in many other countries and is the most effective. It is the fastest expanding and most extensive growing strategy and the second largest based on people initiated on treatment and population coverage, respectively.¹ The standard treatment plan for tuberculosis usually is six months; however, for resistant cases, it may even extend to 2 years with less potent and less toxic drugs¹ with treatment change based on the drug resistance.

The first line of drugs includes ethambutol, isoniazid, pyrazinamide, and rifampicin, and these are part of DOTS (Directly Observed Treatment Short course). Despite the increased availability of medical treatment, new challenges have developed with the growing incidences of drug-resistant tuberculosis that include DR-TB (drug-resistant), MDR-TB (multidrug-resistant), and XDR-TB (extensively drug-resistant).⁶

ATT-associated ocular toxicity was first established in the 1960s.⁷ Ethambutol and linezolid are the most common anti-TB drugs responsible for ocular side effects.^4,8 Rare cases of Isoniazid-induced toxic optic neuropathy have also been reported.^9,10 The toxic neuropathies associated with Ethambutol and linezolid account for about 1% and normally occur after 4-6 months of treatment.¹¹ Optic nerve head and retinal nerve fiber layer changes are minimal in the early stages.⁷ The most common visual field defects associated with ethambutol and linezolid are central and centro-caecal scotomas,¹² whereas bitemporal hemianopic scotomas are more commonly seen with isoniazid toxicity.⁹

Timely detection of ATT-induced optic neuropathy is essential since sight-threatening complications can be reversed if the drug causing the toxicity is discontinued. The purpose of this study was to evaluate visual function in patients receiving ATT and to evaluate the visual field changes caused due to ATT.

Methods

This was a prospective observational study done in a tertiary health care center in South India, conducted from December 2020 to October 2022. The study was approved by the Institutional Ethics Committee (Kasturba Medical College, Mangalore, affiliated to Manipal Academy of Higher Education, Manipal, India, Approval No: IEC KMC MLR 12-2020/441) and abided by the tenets of the Helsinki Protocol. Inclusion criteria were patients above 15 years of age, newly diagnosed with TB who were to be started on anti-tubercular therapy. A written informed consent was taken from all the patients. Patients above 50 years of age, previously on Anti-Tubercular Therapy, patients with a previous history of retinal and optic nerve pathologies, diabetics, patients with renal dysfunction, those with previously documented visual field defects, patients with nutritional optic neuropathy, including vitamin B12 deficiency, and patients with TB arachnoiditis were excluded. Individuals over the age of 50 were excluded from the study, as the risk of glaucoma increases with age and may result in alterations to the visual field.

The following tests were done at baseline (beginning of ATT) and at 6 months from the date of starting ATT. Baseline ocular examination which included best corrected visual acuity using Snellen chart and Jaeger chart, color vision using Ishihara plates (38 plates, 2021 edition), Slit lamp examination, IOP with Goldman Applanation Tonometer, Amsler grid, Dilated fundus examination using slit lamp biomicroscopy and indirect ophthalmoscopy, and Humphrey visual field analysis 30-2 were done. Patients were advised to continue with their ATT and to review immediately in case of any visual symptoms. The occurrence of any visual disturbances supported by diminished visual acuity, colour vision disturbances with or without visual field defects, was considered as toxic optic neuropathy. Visual evoked potential testing (pattern reversal VEP, black and white checkerboard pattern) was done to prove the diagnosis in patients suspected to have toxic optic neuropathy, and the drug responsible was identified and stopped.

The data was analyzed using IBM SPSS version 25. The nominal variables-type of color vision and fields at baseline and follow-up were compared using the Mc Nemer test for nominal variables, whereas the visual acuity was compared at baseline and follow-up using the Wilcoxon sign rank test for ordinal variables. A p-value of <0.05 was considered significant for all analyses.

Results

A total of 48 patients participated in this study. 7 patients who were above the age of 50 years were excluded from the study. The mean age of the study population was 35.90 years ± 10.2 years. Among the 48 patients, 4 (8.3%) were 15-20 years of age, 13 (27%) in the age group of 21-30 years, and 8 (16.6%) in the age group of 31-40. Rest 23 (47.9%) patients are in the age group of 41-50 years.35 (72.9%) out of 48 were male and 13 (27.1%) were female.

Of the total 48 patients included, 32(66.6%) patients had a diagnosis of pulmonary tuberculosis. Out of the 15 patients with extrapulmonary tuberculosis 4(8.3%) had disseminated tuberculosis, 4(8.3%) with TB lymphadenitis, 6(12.5%) with pleural effusion, 1(2.1%) with abdominal tuberculosis and 1(2.1%) with pott’s spine ( Figure 1).

Figure 1. Pi chart showing various indications in the study population for receiving ATT.

TB REGIMEN: MDR tuberculosis, mainly rifampicin-resistant tuberculosis was found in 27(56.3%) at the time of their diagnosis and the rest 21 (43.8%) were drug-sensitive. The average dosage of ethambutol received by the patients in our study was 20 mg/kg body weight for 6 months. The average dosage of INH(Isoniazid) was 5 mg/kg body weight for 6 months. Of the 27 with drug resistance, 25 patients were initiated on a shorter bedaquiline regimen and 3 on all oral longer MDR TB regimens ( Table 1). For the longer MDR TB regimen patients received Linezolid 600 mg for 6 months.

Visual acuity at baseline for all the 48 patients included in the study was 6/6. 3(6.3%) out of 45 patients presented with visual complaints at 6 months. 1 patient (2.1%) had vision of 6/18 with the other two patients (4.2%) below 6/60 ( Table 2).

Color vision examination was normal in all 48 patients during baseline evaluation. 3(6.3%) developed color vision abnormalities at follow-up with all of them being red-green deficient ( Table 3). P value was 0.250 which is not significant.

Anterior segment examination findings and IOP were normal in all patients at baseline and follow-up. Fundus examination was found to be within normal limits in all 48 patients at baseline. However, 3(6.3%) had abnormal findings at follow-up. 1(2.1%) had bilateral temporal disc pallor with the other two patients (4.2%) having bilateral hyperaemic disc.

Visual fields at baseline for all 48 patients were normal. Out of the 45 patients with no visual complaints at 6-month follow-up, 4 (8.33%) showed peripheral field constriction. Out of the 8.33%, Constriction in one quadrant was observed in 2.1% whereas constriction in 2 quadrants was observed in 6.3%. 3(6.3%) patients developed visual symptoms during the treatment course, out of them, peripheral field constriction was noted in one patient (2.1%), one showed severe generalized depression of visual fields (2.1%) and the other patient had central scotoma (2.1%). p-value of 0.16 was calculated which was found to be significant ( Table 4).

All three of these patients with visual complaints were on linezolid as part of their all-oral long bedaquiline regimen. The three patients with visual symptoms were immediately advised to stop the linezolid and even the TB center had been notified of the same.

To confirm the diagnosis in the three patients with visual symptoms VEP and RNFL OCT was done which showed prolonged P100 latency and thinning of the RNFL layer. RNFL OCT and VEP were not done in the patients with only visual field abnormalities without any symptoms.

The visual acuity, color vision, and visual field defects improved in two of the three patients within 1-2 months with the average being 1.5 months post-stopping linezolid. Visual symptoms of one patient however did not improve even after stopping linezolid which can be attributed to a longer duration of drug intake.

Discussion

Baseline visual acuity was normal in all patients; however, 3 (6.3%) patients complained of visual symptoms with 2 reported vision less than 6/60. This is similar to Garg et al,¹³ at 8.69%, Ashraf et al,¹⁴ at 10.6%, Panchal et al,⁴ 7.44%, and Goyal et al,¹⁵ at 3.3% incidence of visual complaints. However, no patient in studies done by Saxena et al,¹⁶ Menon et al,¹⁷ Kandel et al,¹⁸ Kim et al,¹⁹ Jin et al,⁸ had any visual complaints.

Color vision abnormality was noted in 6.3% of our patients similar to 5.32% as seen in a study done by Panchal et al.⁴ All patients in our study exhibited red-green color vision deficiency, as the Ishihara plates used do not detect blue-yellow color vision anomalies.

Out of 48 people in our study on fundus examination, 4.2% had disc edema and 2.1% had temporal disc pallor. The findings in these patients were bilateral. Similar findings were noted by Panchal et al⁴ who had 2.12% with bilateral disc edema and 2.12% with bilateral disc pallor. A study done by Ambika et al,²⁰ showed 44.14% with disc pallor and <1% showed disc edema.

The incidence of visual field defects in our study was 14.6% with the value being significant (0.016). 2.1% had a peripheral constriction in all 4 quadrants, 2.1% with Severe generalized depression of visual fields, and 2.1% with central scotoma. In our study, 4 patients (8.3%) showed Peripheral field constriction in different quadrants making it the most common field defect similar to the study done by Ashraf et al,¹⁴ in which 13.3% had field defects with 8.15% being peripheral field defects in different quadrants. However, visual field defects in patients receiving anti-tubercular therapy varied in different studies. Bitemporal field defects were observed by Kho et al¹² in his study, centro-caecal scotomas (2.12%) were observed by Panchal et al.⁴ Garg et al¹³ in their study concluded that 8.69% (8 eyes of 4 patients) with visual field defects with centro-caecal scotoma in one patient and the rest with peripheral constriction. Few studies like Kandel et al,¹⁸ Kim et al,¹⁹ Jin et al,⁸ Saxena et al,¹⁶ observed no visual field changes in any of their patients on anti-tubercular therapy. The higher incidence of MDR TB in our cohort could be because the study location is the District TB Center and referral center for drug resistant tuberculosis. The limitations of this study include that reaching a definite conclusion by extrapolation of results to the general population was not possible due to the small sample size. The patients were not followed up at shorter intervals, the accurate time of onset of the side effects could not be studied.

Visual field defects caused by ATT can precede clinical symptoms. In our study only 6.3% of the patients presented with decreased vision, whereas 14.6% of the patients had subclinical visual field defects. Subclinical visual field abnormalities could be the first signs of impending optic neuropathy in patients receiving ATT. Visual field evaluation can be an important tool for the early detection of optic neuropathy in patients receiving ATT in clinical settings where visual evoked potential testing and RNFL-OCT are not available. Patients on ATT should be regularly screened for ocular adverse effects preferably every month. Routine examinations like visual acuity, color vision, and fundus examination must be carried out to look for any subclinical effects of the treatment and monitor their progression.

The limitations of the study are as follows. The study population is heterogeneous since all the newly diagnosed tuberculosis patients were included along with those having MDR TB. The patients received the medications according to their body weight. Since fixed drug combinations were prescribed, it is difficult to pinpoint the offending drug. Secondly, we did not compare the Mean deviation and Pattern standard deviation before and after the ATT was started, which could have given us the information regarding cumulative retinal sensitivity.

Chest physicians, ophthalmologists, and healthcare workers need to be aware of potentially sight-threatening side effects of ATT. Prompt diagnosis and timely intervention are the keys to optimal visual outcomes in patients with toxic optic neuropathy due to ATT.

Declarations

Institutional Ethical Committee clearance for the study was obtained.