|
|
| General |
| Study Status |
Completed |
Application Number /
Requirement Number |
P130004 / PAS003 |
| Date Original Protocol Accepted |
05/05/2020
|
| Date Current Protocol Accepted |
|
| Study Name |
Retrospective Analysis Study for Endophthalmitis
|
| Device Name |
RESURE SEALANT
|
| General Study Protocol Parameters |
| Study Design |
Retrospective Cohort Study
|
| Data Source |
External Registry
|
| Comparison Group |
Objective Performance Criterion
|
| Analysis Type |
Analytical
|
| Study Population |
Adult: >21
|
| Detailed Study Protocol Parameters |
| Study Objectives |
The objective of the study is to compare the incidence of endophthalmitis within 30 days of any cataract surgery between sites with and without access to ReSure Sealant using the IRIS registry. The study will also compare the incidence of endophthalmitis within 30 days of cataract surgery between sites with and without access to ReSure Sealant, stratified by type of cataract surgery (i.e., routine and complex).
The study is designed as a retrospective, multicenter, ecologic, post-approval registry study .
|
| Study Population |
The study size will be based on eyes who meet the following inclusion/exclusion criteria:
Inclusion criteria include the following:
1. Presence of either CPT code 66982 or CPT code 66984 code for an extracapsular cataract removal with insertion of intraocular lens prosthesis between January 01, 2016 and December 01, 2019
2. At least one visit in the IRIS database within 30 days after the cataract procedure
3. Practices, with and without access to ReSure, where the cataract procedure with implantation of an IOL was conducted contributing data to the IRIS registry for at least 30 days following the date of the procedure
4. Age 22 and over on the date of the cataract surgery
1. Unspecified laterality of the procedure for cataract removal
2. Unknown demographic information (e.g., age, sex, race and region of the country)
|
| Sample Size |
The minimum sample size needed to detect an increase in the incidence of endophthalmitis in the cohort with access to ReSure Sealant, with a power of 90% and an alpha of 0.05, stratified by estimated incidence of endophthalmitis, versus the cohort without access to ReSure Sealant is 12,806
|
| Key Study Endpoints |
The pre-specified level for clincal significance for the incidence of endophthalmitis in sites with access to ReSure Sealant will be considered to be a doubling of the observed incidence of endophthalmitis in sites without access to ReSure Sealant.
|
| Follow-up Visits and Length of Follow-up |
30 days
|
| Interim or Final Data Summary |
| Actual Number of Patients Enrolled |
In the initial cohort selection, sponsor identified 5,231,758 unique patients and 8,317,635 eyes who had cataract surgery documented in IRIS. After applying inclusion and exclusion criteria, the master cohort contained 4,167,618 unique patients and 6,724,998 eyes from 2,481 practices in the IRIS Registry.
|
| Actual Number of Sites Enrolled |
In this study, 883 practices were considered to have access to ReSure Sealant while 1,598 practices were not for a total of 2,481 study sites.
|
| Patient Follow-up Rate |
30 days post- cataract surgery. Follow-up rate is not applicable because of the study did not involve follow-up of subjects.
|
| Final Safety Findings |
Primary Outcome: “Incidence of endophthalmitis between sites with and without access to ReSure Sealant”:
No clinically significant difference (i.e., doubling) in incidence between eyes treated at practices with ReSure Sealant access (0.609 [95% CI: 0.583, 0.635]) vs. without ReSure Sealant access (0.660 [95% CI: 0.632, 0.688] per 1,000 cataract surgeries (P = 0.01)). Thus, incidence of endophthalmitis within 30 days of cataract surgery is not clinically significantly higher among sites with access to ReSure Sealant as compared to sites without access.
Secondary Outcomes:
o “Incidence of endophthalmitis stratified by cataract surgery type”:
No clinically significant difference in endophthalmitis rate between ReSure access cohort vs. No Access cohort, among eyes that had routine cataract surgeries or complex cataract surgeries:
¿ incidence of endophthalmitis for routine cataract surgeries among ReSure access cohort was 0.572 cases per 1,000 surgeries (95% CI: 0.546, 0.598) and among No Access cohort was 0.614 cases per 1,000 surgeries (95% CI: 0.586, 0.642).
incidence of endophthalmitis for complex cataract surgeries among ReSure access cohort was 1.082 cases per 1,000 surgeries (95% CI: 0.955, 1.209) and among No Access cohort was 1.219 cases per 1,000 surgeries (95% CI: 1.08, 1.357).
o “Incidence of individual endophthalmitis types”: Incidence of secondary infectious iridocyclitis was statistically significantly lower in the cohort with ReSure Sealant access compared with the cohort without access (P < 0.0001). No statistical or clinical significance for unspecified purulent endophthalmitis, panophthalmitis acute or parasitic endophthalmitis .
o “Incidence stratified by covariates”:
No clinically or statistically significant difference in endophthalmitis incidence was found between cohorts with and without ReSure access, stratifying by cataract condition, age group, gender, race, urban/rural, season of surgery, history of diabetes, or concurrent glaucoma and vitreous procedures. Among eyes in West Region, endophthalmitis incidence was 0.490 per 1,000 cataract surgeries (95% CI: 0.441 - 0.540) among cohort with ReSure access, lower than those without access ().
• Other Results: “Risk factors for endophthalmitis”
ReSure Sealant access was not associated with significantly higher risk for endophthalmitis (OR = 0.94, 95% CI = 0.89 - 1.01).
Note: Very large sample sizes may result in statistically significant results that are not clinically meaningful. Therefore, a pre-specified level for clinically meaningful threshold for the observed difference in endophthalmitis rate between sites with and without access to ReSure Sealant was agreed upon with FDA prior to study initiation. This was defined as a doubling of the observed incidence for sites with access compared to sites without access to ReSure Sealant.
|
| Final Effect Findings |
Not applicable for this DER PAS study
|
| Study Strengths & Weaknesses |
Strengths:
The IRIS Registry is the largest electronic health record (EHR)-based comprehensive eye disease and condition registry in the US, a longitudinal dataset that aggregates de-identified patient data from over 232 million patient encounters (53 million unique patients) integrated from more than 55 EHR systems covering 14,945 ophthalmologists and affiliated optometrists (Chang 2019). IRIS Registry extracts data automatically from a participating practice’s EHR system and places in a clinical data repository. In contrast to prospective epidemiological studies using structured protocols, EHR data accrue in an unscheduled, real-world manner.
Weaknesses:
The primary limitation of the research method was the use of the ecologic study design. With this study design, there is no data at the patient level on the exposure to ReSure. The use of ReSure Sealant in the cohort of sites with access to ReSure Sealant was assumed and cannot be verified for a given cataract surgery as use is not routinely coded in EHR medication tables. Since the objective of the research was to identify whether or not use of ReSure Sealant causes an increase risk in endophthalmitis, this design can only be used to identify if the cohort with access to ReSure Sealant has any clinically significant increased risk of endophthalmitis; it cannot predict the impact of access to ReSure Sealant at a patient level.
The second limitation is that study only matched the cohort at the physician level, not the practice level. It was assumed if a physician was matched to the list of NPIs provided by Ocular Therapeutix, then all practices to which that physician belonged were deemed to have access to ReSure Sealant. Patients may not have been exposed to ReSure Sealant if the physician did not use ReSure Sealant at all associated practices, or if the physician did not use ReSure Sealant in all cataract procedures that were performed at the practice. Thus, results cannot show a direct, causal link between endophthalmitis rate and ReSure Sealant use.
Another limitation of the study is the retrospective reliance on the EHR and the physician’s documentation in the EHR. Although this is a legal record, omissions or errors in the documentation may occur and documentation behavior among physicians may differ.
Endophthalmitis reports can also vary over time with spikes in reporting due to various factors including the prevalence of predisposing comorbid conditions, geographic factors in organism incidence, urban versus rural settings, ethnicity, outpatient versus inpatient populations especially those in tertiary care centers. With large sample sizes, these potential variations should be balanced across cohorts, however, it is possible that one or few sites can experience spikes in endophthalmitis thereby impacting the reported incidence for the entire cohort and potentially skewing the study interpretation. Another concern is the misidentification of toxic anterior segment syndrome (TASS), a complication of cataract surgery as endophthalmitis as the two diseases can present with similar symptoms. Incorrect coding of TASS as endophthalmitis might present as several cases in a given site which could skew the results.
|
| Recommendations for Labeling Changes |
Yes
|
|
