Randomized Safety and Feasibility Trial of Ultra-Rapid Cooling Anesthesia for Intravitreal Injections

Written by: Akshay S. Thomas, MD

November 2020

Besirli CG, Smith SJ, Zacks DN, Gardner TW, Pipe KP, Musch DC, Shah AR. Randomized Safety and Feasibility Trial of Ultra-Rapid Cooling Anesthesia for Intravitreal Injections. Ophthalmol Retina. 2020 Oct;4(10):979-986. doi: 10.1016/j.oret.2020.04.001. Epub 2020 Apr 15. PMID: 32446842; PMCID: PMC7541410.

Intravitreal injection is by far the most commonly performed procedure by retina specialists. Lidocaine-based anesthesia prior to intravitreal injections is standard of care, however the medium and duration of anesthetic application varies among retinal specialist. The 2019 ASRS PAT survey found that of its 1004 US-based respondents, 17.9% use an anesthetic-soaked pledget, 22.5% use anesthetic eye drops alone, 25% use anesthetic gel and 33.4% use subconjunctival lidocaine. The preferred duration of anesthetic exposure likely varies from clinician to clinician. Additionally, we have all likely witnessed variations in the degree of discomfort during intravitreal injections between different patients as well as for the same patient from visit to visit. Given that procedural discomfort has been shown to be the leading cause for treatment noncompliance, the availability of an effective, safe, rapid and consistent form of anesthesia prior to intravitreal injections would be of great interest in the retina community.

In this single-center phase 1 dose-ranging study, the authors tested the safety and efficacy of a handheld cooling device for anesthesia prior to intravitreal injections. Twenty-two patients who had been getting bilateral intravitreal injections for wet AMD or DME were recruited and randomized such that 1 eye received standard of care lidocaine-based anesthesia and the other eye received cooling anesthesia. The eye receiving cooling anesthesia was further randomized to 1 of 5 arms which varied in the degree and/or duration of cooling anesthesia (group 1: -5˚C for 10 seconds; group 2: -5˚C for 20 seconds; group 3: -7˚C for 20 seconds; group 4: -10˚C for 10 seconds; group 5: -10˚C for 20 seconds). The standard of care eye was anesthetized with either a series of pledgets or lidocaine gel. Immediately following the injection, patients were asked to rate their pain on a visual analog scale (VAS, range 1-10). Patient were additionally called at time points 4 hours and 7 days following treatment to report subjective pain. Patient were examined 30 minutes and 24 hours following intervention to look for treatment-related adverse effects.

The authors found that a similar proportion of patients in the cooling arm and standard of care arm had either none or mild ocular side effects. Injection-related pain showed a dose-response trend with longer application and colder temperatures achieving greater anesthesia. VAS scores were similar with cooling anesthesia and standard of care immediately following injection for groups 2 through 5 and for all groups at 4 hours following injection. The mean (standard error) VAS pain scores after intravitreal injection was 2.3 (0.4) for the combined standard of care arms and 2.2 (0.6) for patients receiving -10 C cooling anesthesia (groups 4 and 5, p=0.8). Additionally, group 5 experienced a lower immediate post-procedural VAS score than the combined standard of care arms (p=0.02). Finally, the mean (standard error) procedural time for eyes treated with standard of care was 395(40) seconds compared with 124(5) seconds for eyes treated with cooling anesthesia (P < 0.0001).

These phase 1 data show promise for a rapid, safe and reproducible alternative to standard of care lidocaine-based anesthesia for intravitreal injections. The authors note that the -10˚C temperature was warmer than the -20˚C threshold which has been shown to cause reversible histopathologic changes.  It will be interesting to see if cooling anesthesia provides additional prophylaxis against endophthalmitis and if repeated cooling anesthesia generates any significant histopathologic changes.