Management of Suture Penetration in Combined Vitrectomy and Scleral Buckle Surgery

Abstract

Purpose:

To evaluate the surgical outcome of deep or full-thickness suture penetration during combined pars plana vitrectomy and scleral buckle (PPV/SB) surgery.

Methods:

Clinical data of patients who underwent PPV/SB for retinal detachment in our practice between June 2017 and April 2019 were reviewed. Cases with full-thickness scleral penetration were identified. Data including initial presentation, intraoperative complications, and management were collected. Long-term surgical outcome, complications, and recurrent detachment were determined.

Results:

Twenty-three cases with evidence of suture penetration were identified. In none of the cases the penetrated suture was visible in the subretinal space. The suture was replaced in cases with persistent leakage through the penetration site or when the surgical plan involved silicone oil insertion. Subretinal hemorrhage was noted during vitrectomy in 7 (30%) patients. The subretinal hemorrhage migrated to the macula in 2 (9%) cases and was drained through a posterior retinotomy. Retinal incarceration was noted in 4 (17%) cases and was released using a focal retinotomy in 2 (9%)cases to reattach the retina. Late complications included epiretinal membrane in 6 (26%) and recurrent detachment occurred in 2 (9%) patients. Final anatomical success was achieved in all patients.

Conclusions:

We propose new principles in the management of suture penetration during combined PPV/SB surgery. In the absence of a visible penetrated suture, suture replacement is required when there is profuse leakage through the penetration site or if silicone oil is used as tamponade. We recommend removal of submacular hemorrhage through a posterior retinotomy. The incarcerated retina can be flattened using release retinotomies.

Introduction

Inadvertent full-thickness scleral penetration is a well-recognized complication during scleral buckle (SB) surgery. Partial-thickness sutures placed to secure the SB may penetrate the full-thickness of sclera and result in various complications including choroidal or subretinal hemorrhage, retinal perforation, proliferative vitreoretinopathy (PVR), and recurrent retinal detachment. ^{1 -5} Risk factors for scleral penetration include high myopia, scleral thinning, radial buckle placement, hypotony, and reoperation status post-failed buckling procedure. ⁵ Despite the long history and widespread use of SB surgery, data on the incidence, complications, and proper management of suture penetration are limited. Most of our current knowledge about suture penetration is provided by studies of accidental or controlled needle drainage of subretinal fluid during SB surgery. ^2,6 These studies recommend rapid removal of the suture and resuturing of the penetration site to prevent hypotony. ^3,6 If choroidal or subretinal hemorrhage occurs, intraocular pressure is raised with local pressure on the globe with the goal of stopping the bleeding. Occasionally, intraocular gas with postoperative positioning is used to displace submacular hemorrhage. ⁵ Other complications, including retinal incarceration, are usually observed and addressed during future operations if indicated.

In the modern era of retinal detachment surgery, SB surgery is used in combination with pars plana vitrectomy (PPV) for repair of rhegmatogenous retinal detachments in many cases. Vitrectomy techniques have enhanced several aspects of retinal detachment surgery, including management of complications associated with SB surgery. The enhanced visualization during vitrectomy allows proper identification of a full-thickness penetration. Complications such as subretinal hemorrhage and retinal incarceration may be addressed using a variety of techniques and equipment. It is therefore important to revisit some of the current concepts involving surgical complications associated with scleral suture placement and to develop new evidence-based guidelines for their management during combined SB and vitrectomy surgeries.

In this retrospective case series we evaluated the incidence, complications, and management of scleral perforations during combined PPV/SB. We noted a wide range of intraocular findings associated with deep or full-thickness scleral suture penetration. In several cases, it was challenging to establish whether a full-thickness suture passage had indeed occurred. We examined various conventional and new intraoperative approaches to complications of suture penetration and their impact on the surgical and visual outcomes of retinal detachment repair surgeries. Based on our findings, we propose new practical concepts in the management of scleral suture penetrations.

Methods

This study is a retrospective case series on the management of suture penetration during combined PPV/SB. All surgeries were performed by either the corresponding author, or the first author under direct supervision of the corresponding author, during the study period from June 2017 to April 2019.

The surgery begins with a 360° limbal conjunctival peritomy and isolation of the rectus muscles. An encircling band (No. 4050) is passed beneath the rectus muscles and the ends are joined with a silicone sleeve (No. 3084). The band is trimmed to an appropriate length and anchored in each quadrant with a single 5-0 Mersilene (Ethicon Inc) horizontal mattress suture on a spatula needle. The band is tightened with the goal of visibly indenting the retina and supporting 360° of vitreous base, particularly in the quadrant(s) of previously identified tears. Transscleral 23-gauge cannulas are placed through the pars plana. In cases with hypotony secondary to suture penetration, forceps are used to stabilize the globe and facilitate trocar insertion. A complete vitrectomy is performed. All suture sites are closely examined during the vitrectomy for any signs of deep or full-thickness penetration. Endodiathermy is used to demarcate all retinal breaks and to create a posterior retinotomy. The subretinal fluid is drained through the retinotomy. All retinal breaks and the retinotomy are treated with endolaser. Proper indentation and support of the primary break with the SB is the objective. A fluid-air exchange is completed followed by insertion of gas tamponade or silicone oil (SO) in selective cases. The instruments are removed and the sclerotomies and the conjunctiva are closed with 8-0 Vicryl (Ethicon Inc) sutures.

Clinical data, operative reports, and intraoperative videos for patients who underwent combined PPV/SB for retinal detachment were reviewed. Cases with evidence suggesting full-thickness scleral suture penetration were identified. We included patients with sudden leakage of subretinal and/or vitreous fluid during placement of scleral suture with or without transient hypotony, and cases with intraocular findings noted during vitrectomy suggestive of full-thickness suture penetration, including choroidal whitening (choroidal fleck), subretinal hemorrhage, retinal tear/perforation, choroidal hemorrhage, and retinal incarceration at the suture site. Data including initial presentation, preoperative macular status, surgical approach, intraoperative complications and management, postoperative complications, preoperative and postoperative visual acuity (VA) and lens status, duration of follow-up, recurrent detachment, and final anatomical outcome were collected. An optical coherence tomography (OCT) of the macula was obtained in most cases during follow-up visits after resolution of the intraocular gas. If an epiretinal membrane (ERM) was identified on exam and/or on OCT, the ERM was graded based on the grading system published by Govetto et al. ⁷

Results

Data from 182 PPV/SB procedures performed between June 2017 and April 2019 were reviewed. Twenty-three cases (12%) with evidence of full-thickness scleral penetration were identified (Table 1). Of these 23 cases, intraocular gas was used as a tamponade for 22 (96%) cases, and SO was used for 1 (4%) case. The average preoperative logarithm of the minimum angle of resolution VA was 0.98 (Snellen equivalent, 20/200), which improved to 0.43 (Snellen, 20/50) postoperatively. Seventy-four percent of patients achieved a Snellen acuity of 20/60 or better. The average postoperative follow-up was 9 months, ranging from 3 to 21 months.

Table 1.

Patients’ Initial Presentations, Intraoperative Complications, Surgical Procedures, and Outcomes.

Case ID	Mac Status	Preoperative Lens Status	Surgical Procedure	Intraoperative Event	Intraoperative Management	Postoperative Complication	VA Initial	VA Final	Postoperative Lens Status	Follow- Up, mo
1	Mac on	2+ NS	PPV/SB/gas	None	EL	ERMa (gr 1)	20/40	20/40	PCIOL	6
2	Mac off	2+ NS	PPV/SB/gas	None	EL	None	HM	20/60	2+ NS	4
3	Mac on	1+ NS	PPV/SB/gas	Retinal incarceration	Focal retinotomy	None	20/40	20/25	PCIOL	10
4	Mac off	PCIOL	PPV/SB/gas	None	EL	None	20/400	20/50	PCIOL	21
5	Mac off	PCIOL	PPV/SB/gas	Retinal incarceration	EL	ERM (gr 2)	HM	20/50	PCIOL	8
6	Mac off	PCIOL	PPV/SB/gas	Peripheral SRH	EL	Focal PVR, no detachment	20/25	20/30	PCIOL	10
7	Mac off	PCIOL	PPV/SB/gas	Retinal incarceration	EL	PVR and recurrent detachment	20/160	20/60	PCIOL	10
8	Mac off	tr NS	PPV/SB/gas	Submacular SRH	EL, SRH drainage	None	CF at 1 foot	20/80	1+ NS	3
9	Mac off	PCIOL	PPV/SB/gas	Peripheral SRH	EL	None	20/200	20/60	PCIOL	9
10	Mac on	2+ NS	PPV/SB/gas	None	EL	ERM (gr 1)	20/25	20/30	2+ NS	5
11	Mac off	1+ NS	PPV/SB/SO	None	EL, suture replacement, SB repositioning	None	20/60	20/125	1+ NS	3
12	Mac on	1+ NS	PPV/SB/gas	Peripheral SRH	EL	ERM (gr 1)	20/60	20/25	PCIOL	16
13	Mac off	1+ NS	PPV/SB/gas	None	EL	Persistent subretinal fluid, ERM (gr 2)	CF at 1 foot	20/400	PCIOL	16
14	Mac off	2+ NS	PPV/SB/gas	Persistent leakage	EL, suture replacement, SB repositioning	None	20/400	20/100	PCIOL	10
15	Mac off	1+ NS	PPV/SB/gas	Retinal incarceration, persistent leakage	EL, suture replacement, focal retinotomy, SB repositioning	None	20/40	20/40	PCIOL	7
16	Mac on	PCIOL	PPV/SB/gas	None	EL	None	20/70	20/40	PCIOL	4
17	Mac off	PCIOL	PPV/SB/gas	Peripheral SRH	EL	ERM (gr 1)	20/160	20/80	PCIOL	6
18	Mac on	2+ NS	PPV/SB/gas	Two penetrations	EL	PVR with recurrent detachment	20/60	20/40	PCIOL	16
19	Mac off	PCIOL	PPV/SB/gas	Submacular SRH	SRH drainage	None	HM	20/25	PCIOL	13
20	Mac off	PCIOL	PPV/SB/gas	None	EL	None	20/400	20/40	PCIOL	9
21	Mac on	PCIOL	PPV/SB/gas	None	EL	None	20/30	20/30	PCIOL	8
22	Mac on	1+ NS	PPV/SB/gas	Retinal tear	EL	ERM (gr 3)	20/20	20/200	3 + NS	7
23	Mac off	PCIOL	PPV/SB/gas	Choroidal hemorrhage, retinal tear, vitreous hemorrhage	EL	None	4/200	20/50	PCIOL	5

Abbreviations: CF, counting fingers vision; EL, endolaser; ERM, epiretinal membrane; HM, hand motion vision; ID, identification; Mac, macula; NS, nuclear sclerotic cataract; PCIOL, posterior chamber intraocular lens; PPV, pars plana vitrectomy; PVR, proliferative vitreoretinopathy; SB, scleral buckle; SO, silicone oil; SRH, subretinal hemorrhage; VA, visual acuity.

^a ERM grading: gr 1: thin ERM with a foveal depression, distinguishable retinal layers; gr 2: absent foveal depression, retinal distortion, retinal layers still are distinct; gr 3: ectopic inner foveal layers, observable retinal layers; gr 4: disrupted macula, distorted and indistinguishable retinal layers.

A wide range of findings associated with deep or full-thickness suture passage was noted. Findings included transient or persistent leakage of intraocular fluid through the penetration site, transient hypotony, internal whitening of the choroid at the suture site (choroidal fleck), and vitreous, subretinal, or choroidal hemorrhage. In none of the cases was the penetrated suture visible in the subretinal or preretinal space during the vitrectomy.

The penetrated suture was replaced in 3 (13%) of these 23 cases (11, 14, and 15). In cases 14 and 15, persistent leakage of intraocular fluid through the penetration site was noted by external examination of the globe during vitrectomy. In both cases, the suture was removed, and the SB was repositioned to cover the penetration site, which stopped the leakage. A new suture was then placed to secure the SB. In case 11, the planned surgery involved insertion of SO as tamponade. Once suture penetration was identified intraoperatively, the suture was removed and the penetration site was closed using a second suture to prevent potential postoperative leakage of SO into the subconjunctival space. The penetrated suture was left in place in the remaining 20 (87%) of these 23 cases, and the penetration site was treated internally with endolaser during vitrectomy. In these 20 cases, the leakage through the penetration site ceased spontaneously during vitrectomy. In all cases, the globe was stable without hypotony at the conclusion of surgery. There were no cases of postoperative hypotony or endophthalmitis in our series.

The most common intraoperative complication of suture penetration was subretinal hemorrhage, which occurred in 7 (30%) patients (Table 2). In 2 (9%) patients (8 and 19), the subretinal hemorrhage migrated to the macula (Figure 1A). The submacular hemorrhage was drained through a posterior retinotomy with a soft-tip cannula (Figure 2 and Supplemental Video). In both cases, the submacular hemorrhage already created a clot and required gentle manipulation to be removed through the retinotomy without damaging the retina. Both cases recovered without significant postoperative complications. Other cases of localized subretinal hemorrhage were observed without intervention and resolved spontaneously. None of these patients complained of peripheral visual field defects during the postoperative period.

Table 2.

Intraoperative and Postoperative Complications of Scleral Suture Penetration.

Complication	Patients, No. (%)
Intraoperative
Subretinal hemorrhage	7 (30)
Extramacular	5 (22)
Submacular	2 (9)
Retinal incarceration	4 (17)
Persistent leakage from penetration site	2 (9)
Retina tear	2 (9)
Choroidal hemorrhage	1 (4)
Postoperative
Epiretinal membrane	6 (26)
Grade 1	4 (22)
Grade 2 or higher	2 (9)
Proliferative vitreoretinopathy	4 (17)
Recurrent retinal detachment	2 (9)
Persistent subfoveal fluid	1 (4)

Figure 1.

Intraoperative complications of scleral suture penetration during scleral buckle surgery. (A) Submacular hemorrhage. (B) Retinal incarceration. (C) Retinal tear without hemorrhage. (D) Vitreous hemorrhage and suprachoroidal hemorrhage.

Figure 2.

(A) Subretinal hemorrhage involving the fovea after supratemporal scleral penetration in patient 19 (intraoperative images inverted for consistency). (B) The subfoveal hemorrhage was removed using a soft-tip cannula through a posterior retinotomy and the penetration site was treated with endolaser. (C) Seven months later, the retina was attached and the patient maintained 20/25 vision.

In 4 (17%) patients (3, 5, 7, and 15), the detached retina became incarcerated into the penetration site (Figure 1B). The incarceration was minimal in 2 (9%) cases (5 and 7) and the retina was attached after subretinal fluid drainage. Patient 7 later developed significant PVR, which led to recurrent retinal detachment. In 2 other cases (3 and 15), retinal incarceration prevented reattachment of the retina. A focal retinotomy was created to release the traction at the penetration site (Figure 3). The retina was attached after subretinal fluid drainage in both cases and the retinotomy site was treated with endolaser. Both cases remained attached on follow-up without developing significant PVR.

Figure 3.

(A) Retinal incarceration after supratemporal scleral penetration and persistent penetration in patient 15 (intraoperative images inverted for consistency). (B) After removal of penetrated suture, a focal retinotomy was created at penetration site using endodiathermy to flatten the retina and the retinotomy was treated with endolaser. (C) Three and a half months later, the retina was attached and the patient maintained 20/40 vision.

In 2 patients (8%), a retinal tear was noted in proximity to the suture placement site (22 and 23). In both cases, the retinal tear was not noted during the preoperative exam and therefore was attributed to suture placement. Further, in both cases, the retinal tear appeared to be a perforation rather than a flap tear. In case 22, the retina was attached at the penetration site without subretinal hemorrhage (Figure 2D). In this case, the retinal tear was treated with endolaser and the suture was left in place. This patient developed a grade 3 ERM postoperatively that was peeled in a second surgery. However, the visual outcome was unfavorable after the second surgery. In case 23, the retinal tear was associated with vitreous and choroidal hemorrhages (Figure 1C). The vitreous hemorrhage was removed during vitrectomy and the choroidal hemorrhage was observed. The patient recovered with good visual outcome.

The most common postoperative complication of suture penetration was ERM, which occurred in 6 (26%) patients. We used the grading system proposed by Govetto et al to grade the ERM. ⁷ The ERM was grade 1 in most cases and was observed without intervention. In 2 cases (13 and 22), a second surgery with PPV and membrane peel was performed. In case 22, the second surgery was offered after a significant ERM developed following suture penetration as described previously. In case 13, the ERM was grade 2; however, there was associated cystoid macular edema as well as subfoveal fluid. The patient was offered a second surgery with membrane peel, which resulted in improvement of subretinal fluid.

The single-operation success rate was 91% (21 of 23) in our series. Two patients (7 and 18) developed recurrent detachment as a result of PVR and both were offered a second operation with PPV, membrane peel, and SO insertion. Both patients maintained good visual outcome following SO removal, and their retina remained attached over an extended follow-up period. Final anatomical success was achieved in all patients.

Conclusions

In our study, we identified evidence of full-thickness suture penetration in 12% of patients undergoing combined PPV/SB. Other studies have reported lower rates of full-thickness scleral perforation up to 6%. ^5,6 The higher rate of scleral penetration in our study may be due to several reasons. First, previous studies had specific inclusion criteria for full-thickness suture penetration. For example, Brown and Chignell reported accidental drainage of subretinal fluid in 6% of cases, which accounts for full-thickness penetration only under a detached retina. ⁶ We, on the other hand, included all cases with an abnormal finding at the penetration site in our study regardless of visualizing the penetrated sutures. We recognize that our broad inclusion criteria may overestimate the rate of full-thickness suture penetration. Second, previous studies of full-thickness penetration were during SB-only surgeries and relied on external visualization with indirect ophthalmoscopy to confirm suture penetration. However, we benefited from significantly higher resolution and magnification during vitrectomy that allowed us to identify any abnormal finding at the suture site. This suggests that previous studies may have overlooked cases with minimal findings secondary to full-thickness suture penetration.

The most significant challenge in our series was making the determination of when a full-thickness suture penetration had indeed occurred. Leakage of subretinal fluid through a suture site has always been an indicator of full-thickness penetration ⁶ ; however, in our series there were several cases in which we observed leakage of subretinal fluid and associated hypotony during suture placement without any evidence of suture penetration or with a small choroidal fleck when examined internally. One could certainly debate that retinal tears, subretinal hemorrhages, and retinal incarcerations are strong indicators of full-thickness scleral penetration. However, none of these findings establish full-thickness scleral perforation in the absence of a visible suture in the subretinal space during vitrectomy and SB surgery. We therefore suggest that complications such as subretinal hemorrhage or leakage of subretinal fluid may occur with deep passage of scleral sutures as well as full-thickness penetrations. One potential explanation for choroidal or subretinal hemorrhage during deep passage of scleral suture is rotation of the blunt end of the needle into the scleral bed as the needle is pulled through the sclera. ⁵ Under such a mechanism despite damage to the choroid or retina, the deep suture may not be visible when examined internally.

Similarly, we found ourselves debating under what circumstances the deep or potentially penetrated sutures need to be replaced. In the absence of a visible suture in the subretinal space, we replaced the penetrated suture in 2 scenarios. The first scenario is that of persistent leakage through the suture site during vitrectomy, indicating a passage between the intraocular and extraocular space where the infusion fluid passed through the retinal break into the subretinal space and then outside the globe through the penetration site. The second scenario consists of SO used as a tamponade. Although we had only one case in which SO was planned as tamponade, we recommend replacing the penetrating suture in these cases to prevent potential leakage of SO through the penetration site and associated inflammatory response. ^{8 -10} In other cases in which leakage resolved during vitrectomy, we left the suture in place and treated the suture site with endolaser. We did not observe any cases of postoperative hypotony or endophthalmitis in this study. However, this may be due to the relatively low number of cases in our series. Larger studies are required to determine the safety of this practice and to establish the indications for suture replacement during SB surgery.

Several strategies may be considered to reduce the risk of suture penetration. It has been suggested that using the operating microscope for SB surgeries significantly reduces the risk of penetration. ¹¹ We perform SB with and without the operating microscope, and sometimes find the microscope more difficult because of stereoscopic hindrance while working around the microscope, especially in the inferior quadrants. Scleral belt loops are alternatives to sutures for securing encircling bands and are suggested to have similar surgical outcomes. ¹² However, they are currently less commonly used, and data on their safety profile and complications are limited.

Suture penetration resulted in subretinal hemorrhage in 30% of cases in our series. Other studies also report subretinal hemorrhage in 22% to 28% after scleral suture penetration. ^2,13 If the hemorrhage occurs under a detached retina, it may migrate to the submacular space. The most appropriate intervention for submacular hemorrhage has been a subject of longstanding debate. Animal studies provide evidence for a damaging effect of subretinal blood due to chemical toxicity and mechanical barrier. ¹⁴ On the other hand, it has been demonstrated that in the absence of other macular pathologies such as choroidal neovascular membranes, thin hemorrhages are well tolerated without any intervention. ¹⁵ Removal of submacular hemorrhages caused during PPV/SB and internal drainage has been advocated by 2 other studies. Rubsamen et al drained the submacular hemorrhage through a posterior retinotomy or the main retinal break in 8 patients and reported good visual outcome in the majority of cases. ¹⁶ Wade et al studied the outcome of PPV and internal drainage for traumatic or postsurgical macular hemorrhages. They reported significantly superior outcomes with earlier intervention. ¹⁷ In our series, we evacuated submacular hemorrhage though a posterior retinotomy in 2 cases and the visual and surgical outcomes were favorable. We took advantage of a planned posterior retinotomy in both cases. However, the decision to remove a submacular hemorrhage in other surgical scenarios in which a retinotomy is not part of the surgical plan is debatable. Displacing the subretinal hemorrhage with perfluorocarbon is an alternative approach; however, the subretinal hemorrhage usually forms a clot very quickly and adheres to the retina. Intraoperative OCT may be very useful in assessing the thickness of the hemorrhage and determining whether intraoperative drainage is necessary.

Retinal incarceration occurred during 4 (17%) surgeries in our series. Aylward et al reported 2 cases of retinal incarceration during needle drainage of subretinal fluid; however, none of the cases were associated with surgical failure. ² In our series, however, incarcerated retina failed to attach after drainage of subretinal fluid in 2 (9%) cases (3 and 15). Various interventions may be considered to release the incarcerated retina in this situation, including teasing the retina out of the incarceration site, external injection of viscoelastic into the perforation site, and/or injection of perfluorocarbon posterior to the incarceration site. Stopa and Toth described a technique in which outside positive fluid pressure was applied through the sclerotomy to restore the incarcerated retina. ¹⁸ We were unable to dislodge the retina from the incarceration site using these techniques. An alternative has been described that involves creation of a focal retinotomy to release traction. ¹⁹ Using this technique, we were able to release the traction in both cases and reattach the retina.

Although the number of cases in our study precludes us from drawing any statistical conclusions on the risk of PVR after suture penetration, the data suggest that full-thickness scleral penetration may increase the risk of PVR even in the absence of a retinal tear. In 1 (4%) case (6) we observed a focal PVR at the site of penetration that remained contained and did not lead to retinal detachment. More extensive PVR resulting in recurrent detachment developed in 2 (9%) other cases; however, we were not able to confirm that the PVR originated from the penetration site in these cases. Both cases were successfully repaired in subsequent surgeries.

Our study is the only review of complications and managements of suture penetration during combined PPV/SB. Major differences exist between our study and previous studies in the management of scleral suture penetration. Previous studies were conducted on SB-only surgeries with limited interventions to manage suture penetration unless a PPV was added to the procedure. ^3,5,6 However, in our case series, vitrectomy was already planned, which allowed improved detection of suture penetration and provided many tools in definite treatment of complications. We attempted to provide several examples of such treatments in our study; however, the retrospective nature of our study and its relatively low number of cases limit the scope of its conclusions. The surgical planning and associated outcomes reviewed here are specific to these cases and may not be generalizable to all SB surgeries.

In conclusion, we propose new concepts in the management of scleral suture penetration during combined PPV/SB. We suggest that in the absence of a visible suture in the subretinal space, the penetrated suture needs to be replaced only when indicated because of persistent leakage or when SO is planned as tamponade. We recommend drainage of the submacular hemorrhage through a posterior retinotomy. Retinal incarcerations that prevent reattachment of the retina may be released using a focal retinotomy.