Table 1 Summary of studies on the treatment of NAAION (2008–2020)

Guerriero (2009) [30]

III

20 treated

10 untreated

Case–control

N/A

LDL apheresis

Short-term improvement (in the 3 months after onset of the disease) in the MD (−11.08 ± 6.51 vs −16.53 ± 10.03, P = 0.03; −17 ± 5.24 vs −14.14 ± 9.42), but no benefit at 6 months

N/A

Modarres (2011) [31]

IV

31

Case series

(Range 3–22 days)

Intravitreal erythropoietin (2000 units)

VA improved in 27 eyes (87%), 20 eyes (64.5%) showed ≥3 lines of visual improvement at 3 months (P < 0.001)

N/A

Rebolleda (2013) [23]

III

10 treated

27 untreated

Case–control

2 weeks

Oral prednisolone (80 mg daily, tapering dose)

No significant difference between the median change in VA (in the treated group median change in LogMAR VA between the initial and 6-month visit was −0.032 (±0.21) [−0.3; 0.04]), in MD (in treated group median change in MD was −0.56 dB (±5.03) [−3.6; 0.7] ), in PSD (in treated group median change in PSD was −0.02 (±1.9) [−1.6; 0.6]) and average loss in RNFL (in the treated group median change in average RNFLT was 150.5 (±64.9) [119.8; 212.7])

Hypertension, diabetes mellitus, hypercholesterolaemia, aspirin use

Saatci (2013) [32]

III

17 total

(16 patients)

Retrospective

(2–15 days)

Intravitreal ranibizumab (0.5 mg)

VA improvement in 14/17 (BCVA 1.45 ± 0.88 vs 0.77 ± 0.70 LogMar at 1st year), disc swelling improvement in 17 (RNFL was 210 ± 38 microns vs 57 ± 18 microns at 1st year)

N/A

Radoi (2014) [33]

III

21 treated

15 untreated

Retrospective

1 month

Intravitreal triamcinolone (4 mg)

At 6 months, higher proportion of patients with improved VA of more than 1 line (> 5 letters) in the injected group [15 patients (71%)] vs the untreated group [2 patients (13%)] (P = 0.0009).

Mean variation of VA letters at 1 month was 13.81 in treated group vs 0.33 in untreated group (P = 0.003)

Mean variation of MD at 1 month was −1.33 (±1.9) in treated group vs 1.77 (±2.5) in untreated group (P = 0.007)

N/A

Rootman (2013) [34]

III

17 treated

8 untreated

Prospective, non-randomised controlled

15 days

Intravitreal bevacizumab (1.25 mg)

No VA or VF or disc swelling improvement (P = 0.3, P = 0.4 and P = 0.1 respectively)

Non-insulin-dependent diabetes, hypertension, smoking

Prokosch (2014) [35]

III

30 treated patients and 30 untreated patients

Prospective, randomised controlled

3 days

Pentoxifylline (PFX)  IV vs Pentoxifylline + fluocortolone (FC)

Change BCVA for PFX + FC patients was 0.11 ± 0.14 after 3 days and 0.21 ± 0.19 after 6 months, while there was no change in the BCVA score in PFX group at either of these time points (mean change BCVA after 3 days 0.0 ± 0.18, after 6 days 0.05 ± 0.19; P < 0.002 and P < 0.001, respectively).

At 6 months after starting therapy MD in PFX patients was −14.4 ± 9.2 and −16.74 ± 3 in PFX + FC patients, comparison of the data showed no significant difference (P < 0.2)

N/A

Kinori (2014) [36]

III

24 treated

24 untreated

Retrospective case–control

14 days

Methylprednisolone (1 g/day)

Mean initial VA was 20/70 (LogMAR 0.54 ± 0.67) in the treated group and 20/69 (LogMAR 0.54 ± 0.49) in the control group (P = 0.8). At the end of follow-up, VA acuity for control group was 20/80 (0.60 in LogMAR) in the treated group and 20/53 (0.42 in LogMAR) in the control group (P = 0.3).

VF showed defects in 2.4 ± 0.8 in the treated group and 2.0 ± 0.6 quadrants in the control group (P = 0.007).

At final visit, quadrant involvement was 2.6 ± 0.9 in treated group and 2.2 ± 0.7 in control group (P = 0.07)

No statistical difference was found between groups at the end of follow-up (P = 0.2).

Diabetes mellitus, dyslipidaemia, hypertension, ischaemic heart disease and smoking

Zhu (2015) [37]

IV

16

Case series

(6–18 days)

Enhanced extracorporeal counterpulsation (EECP)

10 eyes (62.5%) showed improvement of 3 or more Snellen lines. The median LogMAR VA in NAAION eyes was 0.92 [0.32; 1.92] before EECP and was 0.40 [0.22; 0.90] after the last EECP treatment. Significant difference in median change in VA in NAAION eyes between before EECP and after 12 h EECP (P = 0.003).

Median MD before EECP was −15.90 dB (SD 5.81) vs −15.14 dB (SD 5.02) after 12-h EECP (P = 0.049)

Diabetes mellitus, dyslipidaemia, hypertension, history of ischaemic heart disease, smoking, previous cerebrovascular accident, alcoholism

Lyttle (2016) [38]

III

33 treated

26 untreated

Retrospective

15 days

Levodopa (100 mg levodopa / 25 mg carbidopa three times daily)

Among patients with ≤ 20/60 initial VA, treated participants had significant improvement (P < 0.0001) in the mean change from initial to final LogMAR VA of −0.74 ± 0.56 (95% CI, −0.98 to −0.50), while the mean change for the control group of−0.37 ± 1.09 (95% CI estimate, −1.00 to +0.26) was not significant (P = 0.23). A significant difference between groups was observed (P = 0.0086) with19/23 (83%) in the treated group improving and none got worse, compared with 6/14 (43%) in the control group improving while four (29%) worsened.

The treated group had worse mean initial VF MD at −18.9 dB (±8.8) compared to the mean initial visual field MD of controls of −13.3 dB (±8.5) (P = 0.04). The groups were not found to be significantly different (P = 0.23), with the estimated difference in means at follow-up being 1.78 dB (95% CI, −1.19 to +4.74).

The treated group had a mean reduction of 57.1% RNFLT to 73.2 (±33.2) μm, and control eyes had a mean reduction of 62.5% RNFLT to 85.4 (±19.0) μm compared to the initial RNFL thickness, but this was not statistically significant. No significant difference on VF and disc swelling (P = 0.23 and P = 0.75, respectively)

N/A

Sanjari (2016) [39]

IV

13

Case series

14 days

Intravitreal Fasudil (0.025 mg/0.05 mL)

At M3, BCVA improved from 1.69 ± 0.55 LogMAR at baseline to 0.93 ± 0.51 LogMAR (P = 0.004), RNFLT decreased from 173.5 ± 29.28 μm to 62.9 ± 5.97 μm (P = 0.003) and MD values changed from 24.60 ± 3.80 to 20.5 ± 6.50 (P = 0.005)

N/A

Aftab (2016) [40]

IV

24

Case series

4 weeks

Heparin IV / Warfarin PO

Significant VA improvement in 16 (66.7%), worsening in 1 (4%), average improvement was 5.6 LogMAR lines

Hypertension, dyslipidaemia, diabetes mellitus

Pakravan (2016) [21]

III

90 total

30 untreated

Randomised controlled

N/A

Steroid / normobaric oxygen with mask

Mean initial BCVA was 1.02 ± 0.63, 1.05 ± 0.7 and 0.76 ± 0.5 LogMAR in groups 1 (control), 2 (steroid) and 3 (oxygen), respectively (P = 0.293); corresponding values were 0.8 ± 0.45, 0.84 ± 0.45 and 0.58 ± 0.4 at month 1 (P = 0.127, 0.19 and 0.168, respectively). BCVA improved to 0.71 ± 0.46, 0.73 ± 0.36 and 0.59 ± 0.41 LogMAR at the 6-month follow-up point (P = 0.039, 0.048 and 0.195, respectively).

Initial MD was 19.26 ± 7.02, 20.51 ± 4.68 and 19.3 ± 7.17 in groups 1, 2 and 3, respectively (P = 0.6). Corresponding values at month 1 were 20.26 ± 8.52, 19.52 ± 7.08 and 18.3 ± 7.45, (P = 0.6); and at month 6 were 18.42 ± 8.17, 17.66 ± 6.44 and 16.53 ± 6.32, respectively (P = 0.6).

RNFLT at presentation was 166 ± 57, 184 ± 57 and 193 ± 65 μm in groups 1, 2 and 3, respectively (P = 0.2), which decreased to 73 ± 11, 87 ± 26 and 79 ± 19 μm at the final follow-up (all P < 0.001)

Exclusion of patients with diabetes mellitus or poorly controlled hypertension.

Weiss (2017) [41]

IV

10 total

Case series

(1–35 years)

Autologous bone marrow-derived stem cell therapy

VA improvement in bilateral vision in 80% of patients (P = 0.02) with an average of 3.53 Snellen lines. 73.6% of eyes treated gained vision (P = 0.01) and 15.9% remained stable in the post-operative period. The average LogMAR change in treated eyes was a gain of 0.364 (P = 0.008).

N/A

Pakravan (2017) [42]

III

83 treated

30 untreated

Case–control

14 days

IV erythropoietin and steroid /steroid alone

No significant difference between the 3 groups for BCVA (P = 0.8), MD (P = 0.8), RNFLT (P = 0.1) at 6 months

Hypertension, dyslipidaemia, ischaemic heart disease

Exclusion of patients with diabetes mellitus or poorly controlled hypertension

Saxena (2018) [22]

II

19 treated

19 untreated

Double-blind randomised

1 month

Oral steroids (80 mg tapering dose)

Untreated group showed a median baseline BCVA of 0.8 LogMAR (range, 0–2.7 LogMAR), whereas treated group showed a median baseline BCVA of 1 LogMAR (range, 0.5–3 LogMAR; P = 0.16). The final median BCVAs of the untreated group and treated group were 0.6 LogMAR (range, 0–2.7 LogMAR) and 0.5 LogMAR (range, 0.2–1.8 LogMAR), respectively (P = 0.78). Both groups showed a statistically significant improvement in BCVA from baseline during 6-month follow-up (P = 0.01 and P = 0.003 for the untreated and treated groups, respectively); however, the treated group showed a greater change in vision compared with the untreated group.

Superior and inferior RNFLT showed a greater reduction of oedema at the 1-month follow-up visit (P = 0.028 and P = 0.031, respectively) in the treated group compared with the placebo group. There was a greater change in the superior and inferior quadrants (P = 0.03 and P = 0.03, respectively) at the 1-month follow-up visit in the treated group compared with the untreated group.

The percentage change in RNFLT in both groups was statistically similar.

Hypertension, dyslipidaemia, obstructive sleep apnoea

Exclusion of patients with diabetes

Kalabova (2020) [43]

III

55

Retrospective

N/A

IV vasodilators alone or with IV corticosteroids

In the group treated only with IV vasodilator, mean VA at the beginning of NAAION was 0.356, and immediately after the end of therapy 0.439. Thus, VA improved by an average of 0.083. In the group with combined therapy, average VA before treatment was 0.398 and immediately after the end of therapy 0.429, i.e. VA improved by an average of 0.031. No significant difference was found between the groups in the change of VA (P = 0.7).

Hypertension, diabetes mellitus, dyslipidaemia, hypercoagulation states (hyperhomocysteinaemia), collagenosis, nocturnal hypotension, obstructive sleep apnoea syndrome, treatment for erectile dysfunction and smoking

Nikkhah (2020) [44]

II

Group A (systemic erythropoietin): 33

Group B (oral steroids): 32 Group C (control): 32

Randomised controlled

5 days

10,000 units of erythropoietin / 12 h for 3 days (group A)/ oral prednisone 75 mg /24 h tapered off in 6 weeks

55% of patients in group A (systemic erythropoietin) versus 34.3% in group B (oral steroids) and 31.2% in group C (control) had an improvement of at least 3 lines in the best-corrected VA at M6 (P = 0.04).

Exclusion of patients with systemic conditions such as diabetes or uncontrolled high blood pressure

Durbant (2021) [26]

III

41 treated

27 untreated

Retrospective unmasked and non-randomised

N/A

Intravitreal triamcinolone (4 mg/ 0.1 ml)

Higher proportion of patients improved VA by 2 lines or more in the treated group (49%) compared with the untreated group (11%, P = 0.01). Among patients injected before 15 days, the proportion improving by 2 lines or more (55% vs. 11%, respectively, P = 0.01) and by 3 lines or more (45% vs. 11%, respectively, P = 0.03) were significantly higher than in the untreated group.

Visual field improvement was only observed in the subgroup of patients injected within 15 days with a significant improvement of the mean deviation (dB) within 6 months (P = 0.01).

N/A