Introduction: This study aimed to compare anatomical outcomes of air and perfluoropropane gas (C3F8) tamponade in pars plana vitrectomy for the treatment of rhegmatogenous retinal detachment (RRD). Methods: In this retrospective study, data were gathered from 578 patients (578 eyes) with RRD. The follow-up records of all 578 patients that underwent primary vitrectomy for RRD with air or C3F8 were examined and analyzed. Surgical outcomes of the two groups were compared. Results: A total of 342 eyes were treated with air and 236 with C3F8. The mean follow-up period was 37.65 ± 2.33 months. Baseline and preoperative clinical characteristics were similar between groups, but the period to intraocular bubble disappearance (p < 0.0001), intraocular pressure on the first postoperative day (p < 0.0001), number of cases with intraocular pressure >21 mm Hg within 3 days post-surgery (p < 0.0001), and the number with intraocular pressure >21 mm Hg during follow-up (p = 0.0002) differed significantly between groups. Primary reattachment rates for air and C3F8 groups were 95.03% and 95.34%, respectively. Clinical characteristics were similar in those with and without successful reattachment, and the frequency of new or unclosed breaks was similar between the two groups. There was no significant difference in two groups according to the presence or absence of inferior retinal breaks and inferior detached quadrants. Univariate and multivariate logistic regression identified no risk factor for surgical failure. Conclusions: Air showed equivalent effects to C3F8, with a shorter period to intraocular bubble disappearance, less risk of postoperative intraocular hypertension, and less expense.

Rhegmatogenous retinal detachment (RRD) is one of the most common causes of blindness and visual impairment with reported incidence of 6.3–17.9 per 100,000 population [1‒3]. Surgical treatment of RRD including pneumatic retinopexy, scleral buckling, pars plana vitrectomy (PPV), and combination techniques achieves primary retinal attachment rates from 81% to 92% in uncomplicated cases [4]. In recent decades, PPV has remained an important surgical technique for treatment of RRD at various levels of severity. The choice of intraocular tamponade agent (gas or silicone oil) to ensure effective chorioretinal adhesion in PPV has become one of the most important issues in this form of surgery [5].

The major tamponade agents currently available are various gases and silicone oils. The most commonly used gas tamponades include air, perfluoropropane (C3F8), and sulfur hexafluoride (SF6). Air is nonexpansile, while 100% SF6 expands approximately twice over 1–2 days and 100% C3F8 expands approximately four times over 3–4 days [5, 6]. The time to absorption in the vitreous cavity is almost 1 week for air, 2 weeks for 20% SF6, and within 4 weeks for 12% C3F8[5, 7, 8]. Compared with silicon oil tamponade, gases may spontaneously dissipate in several days or weeks, without surgical removal after retinal adhesion. Furthermore, gases have both higher surface tension and higher buoyancy than silicone oils [9, 10]. According to an updated systematic review, C3F8 and silicone oil perform similarly and the use of either C3F8 or standard silicone oil appears reasonable for most patients with RRD associated with proliferative vitreoretinopathy (PVR) [11].

Since the speed of visual rehabilitation after PPV depends largely on the absorption period of the tamponade, air is increasingly used as an intraocular tamponade. With a shorter duration in the vitreous cavity, air may facilitate the recovery of visual acuity and reduce the period of prone positioning. Moreover, easy access and the nonexpansile nature of sterile air may make it a better choice for patients in less developed areas. A retrospective, observational, and consecutive study by Tetsumoto et al. [12] reported that the surgical results of air tamponade were not inferior to 20% SF6 tamponade for RRD. Sinawat et al. [13] conducted a double-blind, randomized controlled study of 126 patients with RRD and demonstrated that air is an acceptable alternative to C3F8 when treating RRD by pneumatic retinopexy. For the treatment of idiopathic macular hole, air tamponade may have an equivalent effect to that of SF6[14].

The healing procedure of RRDs relies on the expansion and buoyancy of gas. The upward buoyancy of the gas facilitates closure of superior breaks. Therefore, one of the inherent problems in the treatment of RRD with PPV and gas tamponade is the difficulty in producing a direct tamponade on inferior retinal breaks and detachments in the inferior quadrants. In a retrospective study of 524 eyes with primary uncomplicated RRD, no significant difference in the overall reattachment rate of PPV was found between air and 20% SF6 tamponade in cases without inferior retinal break, but significantly lower success rate was found with air than SF6 in those with inferior quadrant breaks [15]. In another prospective study of 64 eyes with RRD, the primary or overall anatomic success rates were not significantly different between air and C3F8 tamponade in patients with inferior retinal breaks [16]; however, small sample size may have limited its statistical power.

Our team has more than 10 years of experience in vitrectomy and the diagnosis and treatment of RRD. In vitrectomy, we apply 360° scleral indentation at the retinal ora serrata to minimize residual vitreous on the retinal surface. The aim of the present 5-year retrospective study was to compare the anatomical success rates of air tamponade and C3F8 tamponade after 23-gauge PPV in patients with RRD.

Patients

This study was conducted in accordance with the Declaration of Helsinki, and the retrospective review of data was approved by the Institutional Review Board of Shanghai General Hospital (approval number: SHGH-2022003). Informed consent was obtained from all the patients. Patients were recruited from the wards of the Shanghai General Hospital from January 2011 to December 2016. The inclusion criteria were as follows: (1) RRD patients diagnosed based on indirect ophthalmoscopy and confirmed by B-ultrasonography; (2) had undergone PPV with air or C3F8 tamponade; (3) able to cooperate with postoperative prone positioning; and (4) with 3 years of follow-up records. The exclusion criteria were (1) a history of vitreoretinal surgery; (2) PVR worse than grade C1; (3) RRDs with giant retinal tears; (4) RRDs with macular hole; (5) RRDs with complications related to high myopia such as retinoschisis, retinal pigment epithelium atrophy; (6) RRDs with choroidal detachment; (7) liquid perfluorocarbone was used during PPV; (8) surgical complication (endophthalmitis, anterior chamber hemorrhage, vitreous body suprachoroidal cavity hemorrhage, etc.) requiring further surgery within a defined period; or (9) diagnosis of severe vision-impairing eye disease, such as advanced glaucoma, preretinal proliferative retinopathy, or macular hole, during the follow-up period.

Surgical Technique

Vitrectomy was performed on all patients by one experienced surgeon (author F.C.). Three-port PPV was performed using the Alcon Accurus System (Alcon Laboratories, Fort Worth, TX, USA). Visualization during vitrectomy was achieved using the RESIGHT 700 Fundus Viewing System (Carl Zeiss Meditec AG, Jena, Germany). After central and peripheral vitreous removal with a 23-gauge incision, all the RRD eyes underwent 360° scleral indentation to shave the vitreous base up to the ora serrata, followed by removal of all vitreous traction from retinal tears. Complete fluid-air exchange was performed, and subretinal fluid (SRF) was aspirated using a flute needle. The patients underwent transscleral cryopexy (peripheral retina) and/or laser (posterior retina) to achieve retinopexy. Laser was used to seal retinal breaks as far as possible, but for those located in the peripheral retina and in accessible by laser, transscleral cryopexy was used. The sclerotomy was closed using 8-0 nylon sutures to avoid gas leakage. In the expansile gas group, at the end of the procedure, 0.5 mL–0.8 mL C3F8 was injected into the vitreous cavity through a tuberculin syringe with a short 27-gauge needle. After intravitreal injection, gentle pressure was maintained on the injection site with a 75% alcohol-rinsed cotton tip applicator for at least 5 min until the injection site was definitely closed. Patients were instructed to maintain the prone position for a specified period of time every day until intraocular bubbles disappeared.

Outcome Measures and Data Collection

Successful surgery was defined as the complete disappearance of SRF and anatomical flattening of the entire circumference of the retinal breaks. Considering the single-operation success, these patients were divided into success and failure group. Patients were followed up regularly at 1 day, 3 days, 1 week, and 1, 3, 6, 12, 18, 24, 36, 48, and 60 months. All the ophthalmic examination results were recorded before and after surgery and included the following: (1) slit-lamp examination and both direct and indirect ophthalmoscopy to assess retinal reattachment and hole closure, optic nerve color, and lens opacity; (2) best corrected visual acuity using a Snellen visual acuity chart; (3) intraocular pressure (IOP) using a noncontact tonometer; (4) ocular B-ultrasonography; (5) optical coherence tomography scan of the macular region; and (6) date of disappearance of intraocular bubbles. According to the eyes with inferior breaks or not and with inferior detached quadrants involved or not, subgroup analysis was performed between the success and failure groups. The following factors were included in logistic regression analysis to identify factors associated with surgical outcomes: age, tamponade type (air vs. C3F8), macula-off/on detachments, high myopia, PVR grade, number of retinal breaks, inferior retinal breaks involved, number of detached quadrants, inferior quadrants involved, and duration of symptoms before surgery.

Statistical Analysis

Commercially available SPSS software (version 22.0; SPSS, Inc., Chicago, IL, USA) and GraphPad Prism (version 8.01, GraphPad software Inc., San Diego, CA, USA) were used for the analyses. Categorical variables (e.g., gender) were expressed as numbers and frequencies, and the differences were assessed using Fisher’s exact test. The continuous variables were expressed as the mean ± standard deviation. The Kolmogorov-Smirnov test was used to determine whether continuous variables fit with a normal distribution. The unpaired t-test was used for normally distributed data and the Mann-Whitney U test for abnormal distributed data. Univariate and multivariate logistic regression analysis was used to identify the risk factors for surgical failure. All the tests were two-sided, and p < 0.05 was considered statistically significant.

In total, 578 eyes of 578 RRD patients (304 males and 274 females) were included. The mean follow-up period was 37.65 ± 2.33 months. Of the 578 eyes, 342 were treated with air tamponade and 236 with C3F8 tamponade. The demographic and preoperative clinical characteristics of both groups are shown in Table 1. No statistically significant difference was found between the groups in age, gender, right or left eye treated, duration of symptoms, preoperative visual acuity, preoperative IOP, PVR grade, number of patients with high myopia, retinal breaks, detached retinal quadrants, and lens status before surgery.

Table 1.

Baseline and preoperative clinical characteristics of all patients

CharacteristicsTotalAirC3F8p value
Patients (eyes), n 578 (578) 342 (342) 236 (236) 
Age, M±SD, years 57.35±8.36 57.87±8.59 56.62±7.97 0.0780 
Gender, n 
 Male 304 178 126 0.7506 
 Female 274 164 110 
Eye, n 
 Right 307 180 127 0.7796 
 Left 271 162 109 
Duration of symptoms, M±SD, days 10.98±8.42 11.06±8.50 10.86±8.32 0.6011 
Preoperative visual acuity, M±SD, LogMAR 1.63±1.07 1.59±1.07 1.68±1.08 0.3140 
Preoperative IOP, M±SD, mm Hg 14.87±2.54 15.01±2.53 14.66±2.55 0.1061 
With high myopia, n (%) 120 (20.76) 66 (19.30) 54 (22.88) 0.2965 
Number of detached quadrants, M±SD, n 2.10±0.73 2.12±0.73 2.07±0.73 0.3757 
With inferior quadrant, n (%) 256 (44.29) 157 (45.91) 99 (41.95) 0.3465 
Number of retinal breaks, M±SD, n 2.12±1.26 2.08±1.24 2.18±1.28 0.3184 
With inferior breaks, n (%) 194 (33.56) 120 (35.09) 74 (31.36) 0.3504 
Macular, n 
 On 327 196 131 0.6676 
 Off 251 146 105 
PVR grade, n 
 B 493 292 201 0.9440 
 C1 85 50 35 
Lens status, n 
 Phakic 521 312 209 0.5374 
 Pseudophakic 52 27 25 
 Aphakic 
CharacteristicsTotalAirC3F8p value
Patients (eyes), n 578 (578) 342 (342) 236 (236) 
Age, M±SD, years 57.35±8.36 57.87±8.59 56.62±7.97 0.0780 
Gender, n 
 Male 304 178 126 0.7506 
 Female 274 164 110 
Eye, n 
 Right 307 180 127 0.7796 
 Left 271 162 109 
Duration of symptoms, M±SD, days 10.98±8.42 11.06±8.50 10.86±8.32 0.6011 
Preoperative visual acuity, M±SD, LogMAR 1.63±1.07 1.59±1.07 1.68±1.08 0.3140 
Preoperative IOP, M±SD, mm Hg 14.87±2.54 15.01±2.53 14.66±2.55 0.1061 
With high myopia, n (%) 120 (20.76) 66 (19.30) 54 (22.88) 0.2965 
Number of detached quadrants, M±SD, n 2.10±0.73 2.12±0.73 2.07±0.73 0.3757 
With inferior quadrant, n (%) 256 (44.29) 157 (45.91) 99 (41.95) 0.3465 
Number of retinal breaks, M±SD, n 2.12±1.26 2.08±1.24 2.18±1.28 0.3184 
With inferior breaks, n (%) 194 (33.56) 120 (35.09) 74 (31.36) 0.3504 
Macular, n 
 On 327 196 131 0.6676 
 Off 251 146 105 
PVR grade, n 
 B 493 292 201 0.9440 
 C1 85 50 35 
Lens status, n 
 Phakic 521 312 209 0.5374 
 Pseudophakic 52 27 25 
 Aphakic 

PVR, proliferative vitreoretinopathy; IOP, intraocular pressure; M, mean; SD, standard deviation.

The postoperative clinical characteristics of each group are summarized in Table 2. The two groups did not significantly differ in final visual acuity, final IOP, lens status, whether single or multiple operations were conducted, or final reattachment rate. However, on the first postoperative day, the mean IOP (15.34 mm Hg) of the C3F8 group was significantly higher than (10.22 mm Hg) that of the air group (p < 0.0001). The number of cases with IOP >21 mm Hg within the first 3 days after PPV and during the whole follow-up were both significantly higher in the C3F8 group than the air group (p < 0.0001, p = 0.0002, respectively). The period to intraocular bubble disappearance, which equates to the prone positioning period, for the air group was 8.61 ± 1.00 days, notably shorter than that of the C3F8 group (21.58 ± 1.58 days, p < 0.0001). The single-operation success rates for the air and C3F8 groups were 95.03% and 95.34% (p = 0.8824), respectively; the final-operation success rates increased to 99.12% and 99.15% (p = 0.9697), respectively, and were statistically similar in the two groups.

Table 2.

Postoperative clinical characteristics and complication of all patients

TotalAirC3F8p value
Follow-up period, M±SD, months 37.65±2.33 37.77±2.52 37.48±2.03 0.1458 
Time of intraocular bubble disappearance after PPV, M±SD, days 13.91±6.50 8.61±1.00 21.58±1.58 <0.0001 
Final visual acuity, M±SD, LogMAR 0.69±0.37 0.70±0.37 0.67±0.38 0.2669 
IOP on the first postoperative day, M±SD, mm Hg 12.31±3.81 10.22±2.04 15.34±3.74 <0.0001 
The number of IOP >21 mm Hg within the first 3 days after PPV, n (%) 23 (3.98) 4 (1.17) 19 (8.06) <0.0001 
The number of IOP >21 mm Hg during the whole follow-up, n 43 (7.44) 14 (4.09) 29 (12.29) 0.0002 
IOP at the last visit, mean (SD), mm Hg 14.90±1.92 14.98±1.93 14.80±1.92 0.2689 
Lens status at last follow-up, n 
 Phakic 136 77 59 0.7584 
 Pseudophakic 433 260 173 
 Aphakic 
Primary reattachment, n (%) 550 (95.16) 325 (95.03) 225 (95.34) 0.8824 
New or unclosed breaks in the nearly 5-year follow-up, n (%) 28 (4.84) 17 (4.97) 11 (4.66) 0.8824 
Final reattachment, n (%) 573 (99.13) 339 (99.12) 234 (99.15) 0.9697 
TotalAirC3F8p value
Follow-up period, M±SD, months 37.65±2.33 37.77±2.52 37.48±2.03 0.1458 
Time of intraocular bubble disappearance after PPV, M±SD, days 13.91±6.50 8.61±1.00 21.58±1.58 <0.0001 
Final visual acuity, M±SD, LogMAR 0.69±0.37 0.70±0.37 0.67±0.38 0.2669 
IOP on the first postoperative day, M±SD, mm Hg 12.31±3.81 10.22±2.04 15.34±3.74 <0.0001 
The number of IOP >21 mm Hg within the first 3 days after PPV, n (%) 23 (3.98) 4 (1.17) 19 (8.06) <0.0001 
The number of IOP >21 mm Hg during the whole follow-up, n 43 (7.44) 14 (4.09) 29 (12.29) 0.0002 
IOP at the last visit, mean (SD), mm Hg 14.90±1.92 14.98±1.93 14.80±1.92 0.2689 
Lens status at last follow-up, n 
 Phakic 136 77 59 0.7584 
 Pseudophakic 433 260 173 
 Aphakic 
Primary reattachment, n (%) 550 (95.16) 325 (95.03) 225 (95.34) 0.8824 
New or unclosed breaks in the nearly 5-year follow-up, n (%) 28 (4.84) 17 (4.97) 11 (4.66) 0.8824 
Final reattachment, n (%) 573 (99.13) 339 (99.12) 234 (99.15) 0.9697 

PPV, pars plana vitrectomy; IOP, intraocular pressure; M, mean; SD, standard deviation.

Table 3 shows that tamponade type, age, duration of symptoms, high myopia, macula-off detachments, PVR grade, number of detached quadrants, the presence of inferior detached quadrants, number of retinal breaks, and inferior breaks all showed no significant difference between success and failure after a single operation. The similar results were both obtained by analysis of success and failure within the air (Table 4) and C3F8 groups (Table 5).

Table 3.

Clinical characteristics of all patients according to the surgical results

CharacteristicsSuccessFailurep value
Patients (eyes), n 550 (550) 28 (28) 
Tamponade 
 Air 325 17 0.8460 
 C3F8 225 11 
 Age, M±SD, years 57.37±8.41 57.14±7.41 0.8899 
Duration of symptoms, M±SD, days 10.94±8.38 11.86±9.23 0.5727 
With high myopia, n (%) 116 (21.09) 4 (13.16) 0.3865 
With macula-off detachments, n (%) 241 (43.82) 10 (35.71) 0.3987 
PVR grade C1, n (%) 79 (14.36) 6 (21.43) 0.3032 
Number of detached quadrants, M±SD, n 2.09±0.73 2.29±0.85 0.1704 
With inferior detached quadrants, n (%) 239 (43.45) 17 (60.71) 0.0729 
Number of retinal breaks, M±SD, n 2.10±1.26 2.43±1.14 0.1821 
With inferior breaks, n (%) 181 (32.91) 13 (46.43) 0.1395 
CharacteristicsSuccessFailurep value
Patients (eyes), n 550 (550) 28 (28) 
Tamponade 
 Air 325 17 0.8460 
 C3F8 225 11 
 Age, M±SD, years 57.37±8.41 57.14±7.41 0.8899 
Duration of symptoms, M±SD, days 10.94±8.38 11.86±9.23 0.5727 
With high myopia, n (%) 116 (21.09) 4 (13.16) 0.3865 
With macula-off detachments, n (%) 241 (43.82) 10 (35.71) 0.3987 
PVR grade C1, n (%) 79 (14.36) 6 (21.43) 0.3032 
Number of detached quadrants, M±SD, n 2.09±0.73 2.29±0.85 0.1704 
With inferior detached quadrants, n (%) 239 (43.45) 17 (60.71) 0.0729 
Number of retinal breaks, M±SD, n 2.10±1.26 2.43±1.14 0.1821 
With inferior breaks, n (%) 181 (32.91) 13 (46.43) 0.1395 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 4.

Clinical characteristics of air group patients according to the surgical results

CharacteristicsSuccessFailurep
Patients (eyes), n 325 (325) 17 (17) 
Age, M±SD, years 57.80±8.64 59.12±7.61 0.5384 
Duration of symptoms, M±SD, days 10.99±8.48 12.41±8.89 0.5022 
With high myopia, n (%) 64 (19.69) 2 (11.76) 0.4194 
With macula-off detachments, n (%) 140 (43.08) 6 (35.29) 0.5271 
PVR grade C1, n (%) 45 (13.85) 5 (29.41) 0.0766 
Number of detached quadrants, M±SD, n 2.11±0.73 2.29±0.77 0.1509 
With inferior detached quadrants, n (%) 147 (45.23) 10 (58.82) 0.2729 
Number of retinal breaks, M±SD, n 2.06±1.25 2.35±1.11 0.3464 
With inferior breaks, n (%) 112 (34.46) 8 (47.06) 0.2887 
CharacteristicsSuccessFailurep
Patients (eyes), n 325 (325) 17 (17) 
Age, M±SD, years 57.80±8.64 59.12±7.61 0.5384 
Duration of symptoms, M±SD, days 10.99±8.48 12.41±8.89 0.5022 
With high myopia, n (%) 64 (19.69) 2 (11.76) 0.4194 
With macula-off detachments, n (%) 140 (43.08) 6 (35.29) 0.5271 
PVR grade C1, n (%) 45 (13.85) 5 (29.41) 0.0766 
Number of detached quadrants, M±SD, n 2.11±0.73 2.29±0.77 0.1509 
With inferior detached quadrants, n (%) 147 (45.23) 10 (58.82) 0.2729 
Number of retinal breaks, M±SD, n 2.06±1.25 2.35±1.11 0.3464 
With inferior breaks, n (%) 112 (34.46) 8 (47.06) 0.2887 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 5.

Clinical characteristics of C3F8 group patients according to the surgical results

CharacteristicsSuccessFailurep value
Patients (eyes), n 225 (225) 11 (11) 
Age, M±SD, years 56.74±8.04 54.09±6.20 0.2825 
Duration of symptoms, M±SD, days 10.86±8.25 11.00±10.12 0.2887 
With high myopia, n (%) 52 (23.11) 2 (18.18) 0.7039 
With macula-off detachments, n (%) 101 (44.89) 4 (36.36) 0.5785 
PVR grade C1, n (%) 34 (15.11) 1 (9.09) 0.5833 
Number of detached quadrants, M±SD, n 2.05±0.72 2.27±1.01 0.3446 
With inferior detached quadrants, n (%) 92 (40.89) 7 (63.64) 0.9560 
Number of retinal breaks, M±SD, n 2.16±1.28 2.55±1.21 0.3350 
With inferior breaks, n (%) 69 (30.67) 5 (45.45) 0.3020 
CharacteristicsSuccessFailurep value
Patients (eyes), n 225 (225) 11 (11) 
Age, M±SD, years 56.74±8.04 54.09±6.20 0.2825 
Duration of symptoms, M±SD, days 10.86±8.25 11.00±10.12 0.2887 
With high myopia, n (%) 52 (23.11) 2 (18.18) 0.7039 
With macula-off detachments, n (%) 101 (44.89) 4 (36.36) 0.5785 
PVR grade C1, n (%) 34 (15.11) 1 (9.09) 0.5833 
Number of detached quadrants, M±SD, n 2.05±0.72 2.27±1.01 0.3446 
With inferior detached quadrants, n (%) 92 (40.89) 7 (63.64) 0.9560 
Number of retinal breaks, M±SD, n 2.16±1.28 2.55±1.21 0.3350 
With inferior breaks, n (%) 69 (30.67) 5 (45.45) 0.3020 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

According to the eyes with inferior breaks or not, no risk factors for surgery failure were identified in patients with or without inferior breaks (Table 6 and Table 7). Similarly, no significant difference was found between two groups no matter what the inferior detached quadrants involved (Table 8 and Table 9). Both univariate and multivariate analyses revealed no significant risk factor for surgical results (Table 10 and Table 11). For those eyes with inferior breaks, the success rate between air and C3F8 groups all had no significant difference according to the specific number of detached quadrants (Table 12).

Table 6.

Sub-analysis of patients with inferior breaks

CharacteristicsSuccessFailurep value
Patients, n 181 13 
Tamponade 
 Air 112 0.9806 
 C3F8 69 
Age, M±SD, years 55.97±8.54 54.23±6.26 0.4685 
Duration of symptoms, M±SD, days 10.14±7.39 8.77±6.71 0.5173 
With high myopia, n (%) 39 (21.55) 3 (23.08) 0.8971 
With macula-off detachments, n (%) 78 (43.09) 3 (23.08) 0.1575 
PVR grade B, n (%) 157 (86.74) 9 (69.23) 0.0827 
Number of detached quadrants, M±SD, n 2.07±0.80 2.48±0.88 0.0919 
Number of retinal breaks, M±SD, n 2.66±1.48 3.23±0.93 0.1708 
Number of inferior retinal breaks, M±SD, n 1.49±0.80 1.77±1.01 0.2372 
CharacteristicsSuccessFailurep value
Patients, n 181 13 
Tamponade 
 Air 112 0.9806 
 C3F8 69 
Age, M±SD, years 55.97±8.54 54.23±6.26 0.4685 
Duration of symptoms, M±SD, days 10.14±7.39 8.77±6.71 0.5173 
With high myopia, n (%) 39 (21.55) 3 (23.08) 0.8971 
With macula-off detachments, n (%) 78 (43.09) 3 (23.08) 0.1575 
PVR grade B, n (%) 157 (86.74) 9 (69.23) 0.0827 
Number of detached quadrants, M±SD, n 2.07±0.80 2.48±0.88 0.0919 
Number of retinal breaks, M±SD, n 2.66±1.48 3.23±0.93 0.1708 
Number of inferior retinal breaks, M±SD, n 1.49±0.80 1.77±1.01 0.2372 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 7.

Sub-analysis of patients without inferior breaks

CharacteristicsSuccessFailurep value
Patients, n 369 15 
Tamponade 
 Air 213 0.8611 
 C3F8 156 
Age, M±SD, years 58.05±8.32 59.67±7.58 0.4610 
Duration of symptoms, M±SD, days 11.33±8.81 14.53±10.45 0.1710 
With high myopia, n (%) 77 (20.86) 1 (6.67) 0.1802 
With macula-off detachments, n (%) 163 (44.17) 7 (46.67) 0.8489 
PVR grade B, n (%) 314 (85.09) 13 (86.67) 0.8667 
Number of detached quadrants, M±SD, n 2.10±0.69 2.13±0.83 0.8570 
With inferior detached quadrants, n (%) 81 (21.95) 4 (26.67) 0.6663 
Number of retinal breaks, M±SD, n 1.83±1.04 1.73±0.80 0.7161 
CharacteristicsSuccessFailurep value
Patients, n 369 15 
Tamponade 
 Air 213 0.8611 
 C3F8 156 
Age, M±SD, years 58.05±8.32 59.67±7.58 0.4610 
Duration of symptoms, M±SD, days 11.33±8.81 14.53±10.45 0.1710 
With high myopia, n (%) 77 (20.86) 1 (6.67) 0.1802 
With macula-off detachments, n (%) 163 (44.17) 7 (46.67) 0.8489 
PVR grade B, n (%) 314 (85.09) 13 (86.67) 0.8667 
Number of detached quadrants, M±SD, n 2.10±0.69 2.13±0.83 0.8570 
With inferior detached quadrants, n (%) 81 (21.95) 4 (26.67) 0.6663 
Number of retinal breaks, M±SD, n 1.83±1.04 1.73±0.80 0.7161 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 8.

Sub-analysis of patients with inferior detached quadrants

CharacteristicsSuccessFailurep value
Patients, n 239 17 
Tamponade 
 Air 147 10 0.8263 
 C3F8 92 
Age, M±SD, years 56.70±8.29 56.00±6.58 0.7329 
Duration of symptoms, M±SD, days 11.30±8.36 10.59±8.68 0.7349 
With high myopia, n (%) 55 (23.01) 3 (17.65) 0.6096 
With macula-off detachments, n (%) 100 (41.84) 4 (23.53) 0.1374 
PVR grade B, n (%) 207 (86.61) 13 (76.47) 0.2452 
Number of detached quadrants, M±SD, n 2.37±0.90 2.41±0.94 0.8472 
Number of retinal breaks, M±SD, n 2.44±1.51 2.94±0.97 0.1785 
With inferior retinal breaks, n (%) 158 (66.11) 13 (76.47) 0.3807 
CharacteristicsSuccessFailurep value
Patients, n 239 17 
Tamponade 
 Air 147 10 0.8263 
 C3F8 92 
Age, M±SD, years 56.70±8.29 56.00±6.58 0.7329 
Duration of symptoms, M±SD, days 11.30±8.36 10.59±8.68 0.7349 
With high myopia, n (%) 55 (23.01) 3 (17.65) 0.6096 
With macula-off detachments, n (%) 100 (41.84) 4 (23.53) 0.1374 
PVR grade B, n (%) 207 (86.61) 13 (76.47) 0.2452 
Number of detached quadrants, M±SD, n 2.37±0.90 2.41±0.94 0.8472 
Number of retinal breaks, M±SD, n 2.44±1.51 2.94±0.97 0.1785 
With inferior retinal breaks, n (%) 158 (66.11) 13 (76.47) 0.3807 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 9.

Sub-analysis of patients without inferior detached quadrants

CharacteristicsSuccessFailurep value
Patients, n 311 11 
Tamponade 
 Air 178 0.6730 
 C3F8 133 
Age, M±SD, years 57.88±8.48 58.91±8.56 0.6921 
Duration of symptoms, M±SD, days 10.66±8.40 13.82±10.14 0.2238 
With high myopia, n (%) 61 (19.62) 1 (9.10) 0.4046 
With macula-off detachments, n (%) 141 (45.34) 6 (54.55) 0.5468 
PVR grade B, n (%) 264 (84.89) 9 (81.82) 0.7806 
Number of detached quadrants, M±SD, n 1.88±0.46 2.09±0.70 0.1393 
Number of retinal breaks, M±SD, n 1.85±0.95 1.64±0.92 0.4750 
CharacteristicsSuccessFailurep value
Patients, n 311 11 
Tamponade 
 Air 178 0.6730 
 C3F8 133 
Age, M±SD, years 57.88±8.48 58.91±8.56 0.6921 
Duration of symptoms, M±SD, days 10.66±8.40 13.82±10.14 0.2238 
With high myopia, n (%) 61 (19.62) 1 (9.10) 0.4046 
With macula-off detachments, n (%) 141 (45.34) 6 (54.55) 0.5468 
PVR grade B, n (%) 264 (84.89) 9 (81.82) 0.7806 
Number of detached quadrants, M±SD, n 1.88±0.46 2.09±0.70 0.1393 
Number of retinal breaks, M±SD, n 1.85±0.95 1.64±0.92 0.4750 

PVR, proliferative vitreoretinopathy; M, mean; SD, standard deviation.

Table 10.

Univariate logistic regression analysis of risk factors for surgery success in subgroup

VariableOR95% CIp value
Age 1.003 0.959–1.049 0.890 
Tamponade (air vs. C3F81.070 0.492–2.328 0.865 
With macula-off detachments (yes vs. no) 1.404 0.636–3.097 0.401 
With high myopia (yes vs. no) 1.709 0.582–5.025 0.329 
PVR (B vs. C1) 0.615 0.242–1.564 0.307 
Number of retinal breaks 1.190 0.920–1.541 0.185 
With inferior retinal breaks (yes vs. no) 1.767 0.823–3.792 0.144 
Number of detached quadrants involved 1.393 0.867–2.237 0.171 
Inferior quadrants involved (yes vs. no) 2.011 0.925–4.374 0.078 
Duration of symptoms 0.988 0.947–1.031 0.573 
VariableOR95% CIp value
Age 1.003 0.959–1.049 0.890 
Tamponade (air vs. C3F81.070 0.492–2.328 0.865 
With macula-off detachments (yes vs. no) 1.404 0.636–3.097 0.401 
With high myopia (yes vs. no) 1.709 0.582–5.025 0.329 
PVR (B vs. C1) 0.615 0.242–1.564 0.307 
Number of retinal breaks 1.190 0.920–1.541 0.185 
With inferior retinal breaks (yes vs. no) 1.767 0.823–3.792 0.144 
Number of detached quadrants involved 1.393 0.867–2.237 0.171 
Inferior quadrants involved (yes vs. no) 2.011 0.925–4.374 0.078 
Duration of symptoms 0.988 0.947–1.031 0.573 

PVR, proliferative vitreoretinopathy; CI, confidence interval; OR, odds ratio.

Table 11.

Multivariate logistic regression analysis of risk factors for surgery success in subgroup with inferior breaks

VariableOR95% CIp value
Age 1.006 0.959–1.056 0.793 
Tamponade (air vs. C3F81.032 0.468–2.272 0.909 
With macula-off detachments (yes vs. no) 1.301 0.582–2.907 0.522 
With high myopia (yes vs. no) 1.862 0.612–5.650 0.274 
PVR (B vs. C1) 0.585 0.227–1.505 0.266 
Number of retinal breaks 1.125 0.844–1.499 0.421 
With inferior retinal breaks (yes vs. no) 1.259 0.441–3.595 0.667 
Number of detached quadrants involved 1.206 0.870–2.062 0.494 
Inferior quadrants involved (yes vs. no) 1.463 0.488–4.382 0.497 
Duration of symptoms 0.989 0.946–1.034 0.640 
VariableOR95% CIp value
Age 1.006 0.959–1.056 0.793 
Tamponade (air vs. C3F81.032 0.468–2.272 0.909 
With macula-off detachments (yes vs. no) 1.301 0.582–2.907 0.522 
With high myopia (yes vs. no) 1.862 0.612–5.650 0.274 
PVR (B vs. C1) 0.585 0.227–1.505 0.266 
Number of retinal breaks 1.125 0.844–1.499 0.421 
With inferior retinal breaks (yes vs. no) 1.259 0.441–3.595 0.667 
Number of detached quadrants involved 1.206 0.870–2.062 0.494 
Inferior quadrants involved (yes vs. no) 1.463 0.488–4.382 0.497 
Duration of symptoms 0.989 0.946–1.034 0.640 

PVR, proliferative vitreoretinopathy; CI, confidence interval; OR, odds ratio.

Table 12.

Comparison of success rate with inferior breaks in specific number of detached quadrants between air and C3F8 groups

Number of detached quadrantsAir (success/failure)C3F8 (success/failure)p value
7/1 5/1 0.8255 
20/2 5/1 0.5949 
59/5 46/2 0.4302 
26/0 12/1 0.1519 
Number of detached quadrantsAir (success/failure)C3F8 (success/failure)p value
7/1 5/1 0.8255 
20/2 5/1 0.5949 
59/5 46/2 0.4302 
26/0 12/1 0.1519 

In the 28 cases without primary retina reattachment, 20 had new retinal breaks, five had unclosed breaks, and 3 patients had macular holes (Fig. 1a); all showed no significant difference between air and C3F8 tamponade (Fig. 1b). Among the 20 patients with new breaks, nine cases had breaks in the thin retina close to the cryopexy scar. After transscleral cryopexy, the local retina would contract over time, causing traction and new breaks in the thin surrounding retinas.

Fig. 1.

Proportion and comparison of three causes for primary reattachment failure. a The proportion of three causes for primary reattachment failure of all 28 patients. b The comparison of the causes for primary reattachment failure between two groups.

Fig. 1.

Proportion and comparison of three causes for primary reattachment failure. a The proportion of three causes for primary reattachment failure of all 28 patients. b The comparison of the causes for primary reattachment failure between two groups.

Close modal

RRD is one of the most common forms of retinal detachment, in which retinal breaks allow the ingress of fluid from the vitreous cavity to the subretinal space, resulting in retinal separation [2]. In RD surgery, tamponade agents are used to provide surface tension across retinal breaks, preventing further fluid flow into the subretinal space until the retinopexy (photocoagulation and/or cryopexy) provides a permanent seal. RRD with PVR often requires surgery to restore normal anatomy and to stabilize or improve vision. PVR usually occurs in association with recurrent RD (after retinal reattachment surgery) but occasionally may be associated with primary RD. In both circumstances, a tamponade agent (gas or silicone oil) is needed during surgery to reduce the rate of postoperative recurrent RD.

Previous reports of single-operation success rates from 73% to 96% [17‒22] are consistent with the present findings. Our results showed that the single-operation and final-operation success rates were almost the same with air or gas tamponade, in agreement with Tetsumoto et al. [12] Another comparative study [15] also showed that the initial anatomic success rates of PPV for RRD were comparable between air and long-acting gas tamponade groups in patients with RRD and superior retinal breaks. However, for those with inferior breaks, it may be more difficult to ensure that retinal breaks remain attached for a sufficient period postoperatively, due to the difficulties of maintaining a strict face-down positioning for effective gas tamponade [12]. RRD with inferior breaks still remains a challenge to intraocular gas tamponade, and it is often noted that air tamponade may be unsuitable for treatment of RRD with inferior retinal breaks.

However, no significant difference in success rate was found in patients with RRD and inferior retinal breaks in the air and C3F8 group. Gravity and buoyancy may facilitate tamponade of breaks by gas within the vitreous cavity, regardless of postoperative position. It is possible that breaks can be sufficiently sealed for an extended duration postoperatively, even when using an air tamponade [12]. Zhou et al. [16] also reported that air had tamponade effects equivalent to C3F8 in treating RRD with inferior breaks. However, they found that the number of involved retinal quadrants was an independent predictor of surgery failure, which is slightly different from the present results. The Scottish Retinal Detachment study included 975 patients with RD and demonstrated that each additional clock hour of detachment was associated with a 13% increased risk of surgery failure [23], supporting the findings of Zhou et al. [16]. In contrast, tamponade effects with air and C3F8 were similar with 1, 2, 3, or 4 detached quadrants, so the number of detached quadrants had no impact on anatomical outcome.

The present results can be explained by the following reasons. First, the incidence of postoperative PVR may have been reduced by 360° scleral indentation at the ora serrata, eliminating residual vitreous. Second, SRF was entirely removed, and retinal breaks were completely sealed using laser or transscleral cryopexy. Third, three scleral puncture ports were closed using 8-0 nylon sutures at the end of each surgery to ensure a degree of IOP on the restored retina. These strategies ensured that air had equivalent tamponade effects to C3F8 in treating RRD, regardless of retinal break location or the number of retinal quadrants detached.

The key to successful retinal reattachment is the removal of vitreous traction around the breaks and the closure of the retinal breaks. In our clinical practice, laser and cryosurgery are two of the most common strategies for sealing retinal breaks. However, cryopexy for peripheral retinal tears could easily lead to the surgically induced scar, increasing the possibility of new tears in the thin surrounding retinas.

In this study, PPV surgery was performed using 23 gauge, and primary anatomical success rates were similar to those of a previous study [22] using 25 gauge. Small-gauge PPV with air tamponade may be effective in treating selected cases of relatively simple primary RRD. Therefore, a large series with long-term follow-up is needed to fully validate this surgical approach.

C3F8, as an expanding gas, may cause postoperative intraocular hypertension. From our observations, the number of cases with IOP >21 mm Hg within the first 3 days after PPV and during the whole follow-up in the C3F8 group were both significantly higher than in the air group. It is necessary to measure IOP regularly after this operation, and if raised, it may be controlled using drugs.

A limitation of this study is its retrospective design. However, the sample size was large and sample selection bias was avoided by one surgeon performing all primary PPVs using similar techniques. Unlike most previous studies, all RRD patients with or without inferior breaks were included. We believe that our study provides valuable insights for RRD treatment of breaks at different retinal locations using 23-gauge PPV, and sub-analysis improves the accuracy of the results. Furthermore, the follow-up period was about 3 years, allowing assessment of long-term prognosis of the treatment.

In summary, in RRD patients with or without inferior breaks or inferior quadrant detachments, air tamponade showed equivalent effects to C3F8, with a shorter period to intraocular bubble disappearance, faster visual recovery, less risk of postoperative intraocular hypertension, and less expense. Further prospective research and a large series from multiple clinical centers with long-term follow-up are needed to confirm these findings and validate this surgical approach.

This study was supported by members of the Department of Ophthalmology in Shanghai General Hospital.

This study was conducted in accordance with the Declaration of Helsinki, and the retrospective review of data was approved by the Institutional Review Board of Shanghai General Hospital, approval number SHGH-2022003. Written informed consent was obtained from each participant before undergoing surgery.

The authors declare no competing interests.

This work was financially supported by the National Natural Science Foundation of China (No. 82171100) and the Special Project for Clinical Research in the Health Industry of Shanghai Health Commission (No. 20214Y0045).

All authors contributed to this work. T.Q., Y.S., and W.W. drafted the manuscript, had full access to all the data in the study, and take responsibility for the integrity of the data and the accuracy of the data analysis. Conception and design: T.Q., Y.M., W.W., and X.X. Data collection: T.Q., Y.S., C.C., K.L., and F.C. Analysis and interpretation: T.Q., Y.S., Y.F., and W.W. Critical revision of the manuscript: T.Q., K.L., Y.F., and W.W. Supervision: T.Q., W.W., and X.X. All authors reviewed and approved the manuscript.

Additional Information

Tianwei Qian and Yan Suo contributed equally to this work.

Data are not publicly available due to ethical reasons. Further inquiries can be directed to the corresponding author.

1.
Mitry
D
,
Charteris
DG
,
Fleck
BW
,
Campbell
H
,
Singh
J
.
The epidemiology of rhegmatogenous retinal detachment: geographical variation and clinical associations
.
Br J Ophthalmol
.
2010 Jun
94
6
678
84
.
2.
Steel
D
.
Retinal detachment
.
BMJ Clin Evid
.
2014 Mar 3
2014
0710
.
3.
Chen
SN
,
Lian
IB
,
Wei
YJ
.
Epidemiology and clinical characteristics of rhegmatogenous retinal detachment in Taiwan
.
Br J Ophthalmol
.
2016 Sep
100
9
1216
20
.
4.
Williamson
TH
,
Lee
EJK
,
Shunmugam
M
.
Characteristics of rhegmatogenous retinal detachment and their relationship to success rates of surgery
.
Retina
.
2014 Jul
34
7
1421
7
.
5.
Vaziri
K
,
Schwartz
SG
,
Kishor
KS
,
Flynn
HW
Jr
.
Tamponade in the surgical management of retinal detachment
.
Clin Ophthalmol
.
2016
;
10
:
471
6
.
6.
Lincoff
H
,
Coleman
J
,
Kreissig
I
,
Richard
G
,
Chang
S
,
Wilcox
LM
.
The perfluorocarbon gases in the treatment of retinal detachment
.
Ophthalmology
.
1983 May
90
5
546
51
.
7.
Thompson
JT
.
Kinetics of intraocular gases. Disappearance of air, sulfur hexafluoride, and perfluoropropane after pars plana vitrectomy
.
Arch Ophthalmol
.
1989 May
107
5
687
91
.
8.
Thompson
JT
.
The absorption of mixtures of air and perfluoropropane after pars plana vitrectomy
.
Arch Ophthalmol
.
1992 Nov
110
11
1594
7
.
9.
Petersen
J
.
The physical and surgical aspects of silicone oil in the vitreous cavity
.
Graefes Arch Clin Exp Ophthalmol
.
1987
;
225
(
6
):
452
6
.
10.
Krzystolik
MG
,
D’Amico
DJ
.
Complications of intraocular tamponade: silicone oil versus intraocular gas
.
Int Ophthalmol Clin
.
2000 Winter
40
1
187
200
.
11.
Schwartz
SG
,
Flynn Jr
HW
,
Wang
X
,
Kuriyan
AE
,
Abariga
SA
,
Lee
WH
.
Tamponade in surgery for retinal detachment associated with proliferative vitreoretinopathy
.
Cochrane Database Syst Rev
.
2020 May 13
2020
Cd006126
.
12.
Tetsumoto
A
,
Imai
H
,
Hayashida
M
,
Otsuka
K
,
Matsumiya
W
,
Miki
A
.
The comparison of the surgical outcome of 27-gauge pars plana vitrectomy for primary rhegmatogenous retinal detachment between air and SF6 gas tamponade
.
Eye (London, England)
.
2020 Feb
34
2
299
306
.
13.
Sinawat
S
,
Ratanapakorn
T
,
Sanguansak
T
,
Prompol
S
,
Laopaiboon
M
,
Yospaiboon
Y
.
Air vs perfluoropropane gas in pneumatic retinopexy: a randomized noninferiority trial
.
Arch Ophthalmol
.
19602010 Oct
128
10
1243
7
.
14.
Hasegawa
Y
,
Hata
Y
,
Mochizuki
Y
,
Arita
R
,
Kawahara
S
,
Kita
T
.
Equivalent tamponade by room air as compared with SF(6) after macular hole surgery
.
Graefes Arch Clin Exp Ophthalmol
.
2009 Nov
247
11
1455
9
.
15.
Tan
HS
,
Oberstein
SYL
,
Mura
M
,
Bijl
HM
.
Air versus gas tamponade in retinal detachment surgery
.
Br J Ophthalmol
.
2013 Jan
97
1
80
2
.
16.
Zhou
C
,
Qiu
Q
,
Zheng
Z
.
Air versus gas tamponade in rhegmatogenous retinal detachment with inferior breaks after 23-GAUGE pars plana vitrectomy: a prospective, randomized comparative interventional study
.
Retina
.
2015 May
35
5
886
91
.
17.
Wickham
L
,
Connor
M
,
Aylward
GW
.
Vitrectomy and gas for inferior break retinal detachments: are the results comparable to vitrectomy, gas, and scleral buckle
.
Br J Ophthalmol
.
2004 Nov
88
11
1376
9
.
18.
Schwartz
SG
,
Flynn
HW
.
Primary retinal detachment: scleral buckle or pars plana vitrectomy
.
Curr Opin Ophthalmol
.
2006 Jun
17
3
245
50
.
19.
Mokete
B
,
Williamson
TH
.
Scleral buckling combined with vitrectomy for the management of rhegmatogenous retinal detachment associated with inferior retinal breaks
.
Eye
.
2009 May
23
5
1233; author reply 1233-4
.
20.
Mehta
S
,
Blinder
KJ
,
Shah
GK
,
Grand
MG
.
Pars plana vitrectomy versus combined pars plana vitrectomy and scleral buckle for primary repair of rhegmatogenous retinal detachment
.
Can J Ophthalmol
.
2011 Jun
46
3
237
41
.
21.
Duvdevan
N
,
Mimouni
M
,
Feigin
E
,
Barak
Y
.
25-GAUGE pars plana vitrectomy and SF6 gas for the repair of primary inferior rhegmatogenous retinal detachment
.
Retina
.
2016 Jun
36
6
1064
9
.
22.
Li
Y
,
Cheung
N
,
Jia
L
,
Zhang
H
,
Liu
N
.
Surgical outcomes OF 25-GAUGE pars plana vitrectomy using air as an internal tamponade for primary rhegmatogenous retinal detachment
Philadelphia (PA)
Retina
2020 Jan 7
.
23.
Mitry
D
,
Awan
MA
,
Borooah
S
,
Siddiqui
MAR
,
Brogan
K
,
Fleck
BW
.
Surgical outcome and risk stratification for primary retinal detachment repair: results from the Scottish Retinal Detachment study
.
Br J Ophthalmol
.
2012 May
96
5
730
4
.