Purpose
: To assess the incidence of unplanned return to the operating room (ROR) at ≤45 days or ≥46 days after primary retinal detachment (RD) surgery and correlate ROR with preoperative risk factors and visual outcomes.
Design
: Retrospective cohort study.
Methods
: This was a retrospective review of patients with primary RD surgery to assess for unplanned ROR between January 1, 2012 and June 30, 2014, with follow-up of 90 days to 8 years (mean, 1.5 years). We assessed 268 patients receiving 270 primary rhegmatogenous RD surgeries between January 1, 2012 and June 30, 2014 in an academic tertiary referral center.
Results
: Of the 270 RD surgeries, 82 were complicated (history of proliferative vitreoretinopathy or trauma-related RDs at presentation) and 188 were uncomplicated (RD unrelated to trauma or proliferative vitreoretinopathy at presentation). The ROR rate for all surgeries was 12.2% (33/270) over the follow-up period, with 51.5% (17/33) having reoperations within 45 days. The complicated detachment group had a ROR rate of 14.6% (12/82) over the follow-up period, and 50% of those (6/12) had reoperations within 45 days. The uncomplicated detachment group had a ROR rate of 11.2% (21/188) over the follow-up period. Of those, 52.4% (11/21) had reoperations within 45 days.
Conclusions
: Given that only 51.5% of all RORs occurred within 45 days, a 45-day ROR surgical quality metric that has been previously used may be of limited value for RD surgery. Factors such as age at presentation, number of retinal breaks, number of detached clock hours, use of silicone oil tamponade for pars plana vitrectomy, history of choroidal detachment, high myopia, ocular trauma, and open globe were associated with increasing risk of ROR. Implementing risk-adjusted metrics may provide a more accurate and useful quality improvement metric for evaluating quality of surgical care in vitreoretinal surgery. Am J Ophthalmol 2021;221:•••–•••. © 2021 Elsevier Inc. All rights reserved.
Quality metrics such as “return to the operating room” (ROR) have become increasingly recognized as a surrogate for surgical adverse events and quality of surgical care delivery. The utilization of unplanned RORs has been suggested as a surrogate for quality of care delivery, and therefore ROR time cutoffs have been controversial in the literature. , Previous studies have noted limitations to this approach by suggesting that while quality measures are relatively easy to evaluate, they can be oversimplified and misleading for any given specialty. Recently, a 45-day ROR metric was proposed to assess postoperative ROR events across all surgical specialties at a tertiary care academic medical center. However, the applicability of this 45-day ROR metric has not been specifically studied for surgical retinal detachment (RD) repair.
Within vitreoretinal surgery, previous studies have examined the incidence of reoperation and factors that lead to ROR after RD surgery. However, timing to ROR after initial RD surgery has been studied on a more limited basis. The purpose of this study is to assess the incidence of unplanned ROR at ≤45 days or ≥46 days after primary RD surgery in order to understand the feasibility of a 45-day ROR metric for RD repair surgery. This study also aims to assess the risk factors that contribute to ROR and visual acuity (VA) outcomes of patients undergoing RD surgery.
Methods
A retrospective review was conducted using the electronic medical record to analyze the postoperative outcomes of primary RD surgeries performed at Mayo Clinic, Rochester, Minnesota, USA between January 1, 2012 and June 30, 2014. Institutional review board approval was obtained; this study was compliant with the Health Insurance Portability and Accountability Act.
Data were extracted using the electronic medical record for 268 patients with 270 RD surgeries performed between January 1, 2012 and June 30, 2014. Eyes with previous retinal cryotherapy and previous retinal laser therapy were included in the study. Eyes that had previous RD surgery, pneumatic retinopexy, diabetic tractional RD, proliferative diabetic retinopathy (PDR), or <90 days of follow-up after RD surgery were excluded. Unless there was a recurrent detachment or complication of the initial procedure that was addressed at the time of oil removal, silicone oil (SO) removals were not considered a ROR event.
The following information was collected from the electronic medical record: patient age at the time of surgery, operated eye, sex, procedure performed, macular status (on vs off), number of retinal breaks, presence of grade C proliferative vitreoretinopathy (PVR), number of clock hours of RD, duration of RD, VA at presentation (using Snellen VA), history of ocular surgery, tamponade used at surgery, presence of giant retinal tear (defined as any tear that extends ≥3 clock hours circumferentially), lens status, presence of lattice degeneration, presence of hypotony (pressure at presentation of <8 mm Hg), presence of choroidal detachment, refractive error, history of ocular trauma, postoperative VA, intraoperative complications, postoperative complications, best-corrected VA (BCVA) before RD surgery, and BCVA at every follow-up visit with a minimum of 45 days after surgery. All patients had a minimum of 90 days of follow-up.
The patients were divided into 3 cohorts: 1) a combined group of both uncomplicated and complicated RDs; 2) complicated RDs; and 3) uncomplicated RDs. Complicated detachments were defined as cases with PVR or trauma-associated RDs, whereas uncomplicated detachments were defined as cases without associated trauma or PVR. Trauma and PVR were chosen as cohort exclusion criteria for this study because this mimics the exclusion criteria for Quality Measures 384 in the Intelligent Research in Sight Registry at the time of conducting this study.
The cases were analyzed for 3 outcome measures: 1) time to ROR; 2) complications requiring ROR at any time during follow-up after the initial RD surgery; and 3) postoperative VA at the most recent follow-up visit. The electronic medical record was reviewed to identify possible causes for each ROR.
Statistical Analysis
The frequency of ROR was estimated using the Kaplan-Meier method. Characteristics potentially associated with ROR were assessed using a proportional hazards regression with hazard ratios (HRs) listed for each risk factor. The analysis of tamponade was limited to patients who had a pars plana vitrectomy (PPV) as their index surgery because the use and type of tamponade used during scleral buckling (SB) procedures is highly variable. Findings from these analyses are summarized by presenting the point estimate and 95% confidence interval for the HR. Patient characteristics are summarized using the mean ± the standard deviation (SD) for continuous variables. Nominal variables were summarized using frequency counts and percentages. All analyses were performed with JMP software (JMP Pro 10; SAS Institute Inc, Cary, North Carolina, USA).
Results
Of the 268 patients that met the inclusion criteria and underwent RD surgery, 82 had complicated detachments and 188 had uncomplicated detachments. The mean follow-up time was 563.8 ± 502.0 days with a median follow-up period of 411.5 days. The mean duration of detachment at the time of surgery was 56.7 ± 287.7 days (range 0-4368 days). The mean ± SD VA at presentation was 0.88 ± 0.78 logarithm of minimal angle of resolution (logMAR; 20/151.7) and the mean ± SD final VA was 0.41 ± 0.50 logMAR (20/51.4). Demographic characteristics of the patients included in the study are shown in Table 1 .
| Total Patients | Complicated Patients | Uncomplicated Patients | |
|---|---|---|---|
| Patients (n) | 270 | 82 | 188 |
| Mean age at presentation (y) ± SD | 58.6 ± 14.8 | 57.6 ± 18.9 | 59.0 ± 12.6 |
| Sex, n (%) | |||
| Female | 108 (40) | 38 (46.3) | 70 (37.2) |
| Male | 162 (60) | 44 (53.7) | 118 (62.8) |
| Procedure, n (%) | |||
| Pars plana vitrectomy | 90 (33.3) | 19 (23.2) | 71 (37.8) |
| Scleral buckle | 84 (31.1) | 24 (29.3) | 60 (31.9) |
| Scleral buckle/pars plana vitrectomy | 96 (35.6) | 39 (47.5) | 57 (30.3) |
| Macula status, n (%) | |||
| On | 117 (43.3) | 24 (29.3) | 93 (49.5) |
| Off | 153 (56.7) | 58 (70.7) | 95 (50.5) |
| Breaks (n), mean ± SD | 2.1 ± 1.7 | 1.9 ± 1.6 | 2.2 ± 1.7 |
| Proliferative vitreoretinopathy grade C, n (%) | 23 (8.5) | 23 (28.0) | 0 (0) |
| Clock hours (n), mean ± SD | 5.2 ± 2.5 | 6.5 ± 2.9 | 4.7 ± 2.0 |
| Duration of detachment (days), mean ± SD | 56.7 ± 287.7 | 110.6 ± 501.9 | 33.2 ± 91.0 |
| Previous cryotherapy, n (%) | 4 (1.5) | 1 (1.2) | 3 (1.6) |
| Previous laser history, n (%) | 34 (12.6) | 10 (12.2) | 24 (12.8) |
| Tamponade a | |||
| Patients (n) | 209 | 65 | 144 |
| Sulfur hexafluoride, n (%) | 150 (71.8) | 31 (47.7) | 119 (82.6) |
| perfluoro-n-octane (PFO), n (%) | 38 (18.2) | 21 (32.3) | 17 (11.8) |
| Silicone oil, n (%) | 21 (10.0) | 12 (20.0) | 8 (5.6) |
| Giant retinal tear, n (%) | 7 (2.6) | 3 (3.7) | 4 (2.1) |
| Lens status, n (%) | |||
| Phakic | 159 (58.%) | 50 (61.0) | 109 (58.0) |
| Pseudophakic | 108 (40.0) | 30 (36.6) | 78 (41.5) |
| Aphakic | 3 (1.1) | 2 (2.4) | 1 (0.5) |
| Lattice degeneration, n (%) | 90 (33.3) | 11 (13.4) | 79 (42.0) |
| Choroidal detachment, n (%) | 7 (2.6) | 7 (8.5) | 0 (0) |
| Spherical equivalent (diopter) | |||
| Patients (n) | 185 | 56 | 129 |
| Spherical equivalent ± SD | −3.4 ± 3.7 | −2.4 ± 3.7 | −3.9 ± 3.6 |
| Trauma-related detachment, n (%) | 20 (7.4) | 20 (24.4) | 0 (0) |
| History of open globe, n (%) | 8 (2.9) | 8 (9.8) | 0 (0) |
a Recorded for pars plana vitrectomy only to standardize a comparison among sulfur hexafluoride, perfluoro-n-octane (PFO), and silicone oil.
The combined group had a 45-day ROR of 6.3% (17/270) and a cumulative ROR rate over the follow-up period of 12.2% (33/270), with 51.5% (17/33) of ROR events occurring within 45 days ( Figure 2 ). A Kaplan-Meier failure estimate of timing to ROR is shown in Figures 1 and 2 . The mean time to ROR in the combined group was 70.9 ± 71.8 days (range 5-312 days). The mean final VA for patients who required ROR within 45 days was 1.17 ± 1.77 logMAR (20/295.8) vs 0.35 ± 0.39 logMAR (20/44.8) in those without ROR. Factors such as PPV requiring SO tamponade (HR 3.74, P = .004), history of choroidal detachment (HR 4.66, P = .037), high myopia (HR 1.14, P = .044), trauma-related detachment (HR 4.55, P = .0013), and history of open globe (HR 11.6, P < .0001) were associated with a statistically significant increased risk of ROR in the combined detachment cohort. Among the patients from the combined detachment group with ROR within 45 days, comorbidities that may have contributed to ROR included recurrent detachment (n = 4), open break (n = 4), open globe history (n = 3), acute retinal necrosis (n = 1), PVR grade C (n = 3), exposed SB (n = 1), and perforation during SB (n = 1).
