Middle East African Journal of Ophthalmology

: 2019  |  Volume : 26  |  Issue : 3  |  Page : 163--167

Efficacy and safety of a novel blunt cannula trans-sub-Tenon's retrobulbar block for vitreoretinal surgery

Tafadzwa Young-Zvasara, Johanna Winder, Sidath Wijetilleka, Laurie Wheeler, Roger Mcpherson 
 Cardiff Eye Unit, University Hospital of Wales, Cardiff, Wales

Correspondence Address:
Dr. Tafadzwa Young-Zvasara
Cardiff Eye Unit, University Hospital of Wales, Heath Park, Cardiff, CF14 4XW


PURPOSE: To evaluate a novel trans-sub-Tenon's retrobulbar block (TSTRB) compared to sub-Tenon's block (STB) and peribulbar block (PBB) anesthesia for vitreoretinal surgery. METHODS: This study was a prospective evaluation of cases undergoing TSTRB, STB, or PBB. The Kallio scale and Brahma scales were used to score hemorrhage and extraocular motility, respectively. Pain was documented on a visual analog score graded (1–10) at induction, intraoperatively, and postoperatively, any confounding variables were noted. RESULTS: Seventy eyes have been used in this analysis, of which TSTRB was used in 37% (n = 26), PBB in 34% (n = 24), and STB in 29% (n = 20). Postoperative analgesia was required by 10% (n = 2) of STB and 8% (n = 2) of PBB; none of the TSTRB cases required analgesia (P = 0.003). The mean volume required with each technique was as follows: TSTRB, 4.8 ml; STB, 5.3 ml; and PBB, 10.4 ml (P = 0.030). The volume of anesthesia was correlated with the level of proptosis and even more important affected the ease of surgery most (P = 0.005). Akinesia was greatest with TSTRB > PBB > STB (P = 0.040). There were no complications such as brainstem anesthesia, globe perforation, or retrobulbar hemorrhage. CONCLUSION: Intentionally extending a STB into the retrobulbar space, via a TSTRB fenestration utilizes a familiar skill set. TSTRB produced the best levels of reduced kinesia during surgery and increased duration of postoperative analgesia. The technique uses a small-volume anesthesia.

How to cite this article:
Young-Zvasara T, Winder J, Wijetilleka S, Wheeler L, Mcpherson R. Efficacy and safety of a novel blunt cannula trans-sub-Tenon's retrobulbar block for vitreoretinal surgery.Middle East Afr J Ophthalmol 2019;26:163-167

How to cite this URL:
Young-Zvasara T, Winder J, Wijetilleka S, Wheeler L, Mcpherson R. Efficacy and safety of a novel blunt cannula trans-sub-Tenon's retrobulbar block for vitreoretinal surgery. Middle East Afr J Ophthalmol [serial online] 2019 [cited 2019 Nov 22 ];26:163-167
Available from: http://www.meajo.org/text.asp?2019/26/3/163/268241

Full Text


Ophthalmic surgery complexity and delivery continues to evolve. A surgeon undertaking vitreoretinal procedures requires a well-anesthetized eye which remains still in the primary position. Anesthetic factors, e.g., pain, residual motility, and volume differentials, can compromise surgery, making even routine vitreoretinal procedures challenging.

Sub-Tenon's block (STB) was described by Turnbull in 1884.[1] It has steadily gained popularity and largely replaced “sharp needle” local anesthesia which carries a higher risk of globe perforation. A United Kingdom national survey of cataract surgery in 2003 showed that STB was used in 43% of procedures (up from 7% in 1996), peribulbar block (PBB) in 31%, and topical anesthesia in 21%.[2] It has been established that during STB, the volume required for adequate anesthesia can result in a rise in intraocular pressure (IOP), reduce the depth of the anterior chamber, or cause conjunctival chemosis, leading to surgical challenges. Volumes above 5 ml have been found to produce significant rises in IOP.[3] Although we are seeing a rise in STB use, no guarantee of adequate anesthesia nor akinesia is possible. Sharp needle blocks (peribulbar and retrobulbar) are therefore still used during vitreoretinal surgery. During vitreoretinal surgery, trivial eye movements can significantly compromise safety when performing minute manipulations close to or on the retinal surface such as retinal membrane peeling.

Avoiding a sharp needle block reduces the danger of globe perforation. We aimed to evaluate a standard STB or PBB to a novel trans-sub-Tenon's retrobulbar block (TSTRB) which enters the retrobulbar space using a blunt cannula technique.


The data were collected prospectively on patients undergoing STB, PBB, or TSTRB for vitreoretinal surgery.

TSTRB technique

The patient was positioned comfortably supine, usually without the need for IV sedation. Standard electrocardiogram monitoring was recommended during the placement of the block. Local anesthetic (LA) drops were administered to the conjunctiva, followed by iodine 5% aqueous drops to the conjunctival sac, hair follicles, and external eyelid. A 5-ml syringe of LA mixture was prepared using 1 ml of lignocaine 2%, 4 ml of bupivacaine (or equivalent) 0.5%, and hyaluronidase 150 IU. A STB with 2 ml of this anesthetic mixture was established by utilization of a conventional method, i.e., a 19G sub-Tenon's cannula (Beaver-Visitec, Waltham, MA, USA) was introduced via an inferonasal cut in the conjunctival mucosa and thereafter hydrodissected to an endpoint well past the equator of the globe, at all times remaining deep to Tenon's fascia. The 19G sub-Tenon's cannula was then substituted for a blunt tipped 26G curved lacrimal cannula (Beaver-Visitec, Waterloo Industrial Estate, Warwickshire, England), which was introduced into the previously established sub-Tenon's entry point. The curve of the lacrimal cannulae should follow the curve of the globe in a congruous way, and due to the previous hydrodissection, should slide easily without snagging. Just before the tip gains the same end point as the previous 19G termination point, the operator rotates the 5 ml syringe 180°, encouraging the tip to both move safely away from globe surface and also pierce the Tenon's fascia to gain entry into the retrobulbar space (formed within the cone of extrinsic muscles of the eye) [Figure 1]. After test withdrawal to avoid intravenous or cerebrospinal fluid placement, a further 2 ml of LA mixture was administered. The upper eyelid is usually seen to close during this injection, and most eye movements have stopped within 60 s. This is taken to be a surrogate sign to confirm retrobulbar extension of the block. If further “top-ups” with LA are required, only the standard 19G sub-Tenon's cannula is required, because LA now leaks to the retrobulbar space via the newly made fenestration in the Tenon's fascia. A standard technique was used for the PBB and STB. Surrogate markers of adequate volume included ptosis of the eyelid and or proptosis.{Figure 1}

To quantify hemorrhage, the Kallio scale was used, assigning the following scores: 1 = spot ecchymosis; 2 = eyelid ecchymosis involving half of the eyelid surface area or less; 3 = eyelid ecchymosis all around the eye with no increase in IOP; and 4 = retrobulbar hemorrhage; use of anticoagulants or antiplatelet agents was noted.[4] Immediate and late complications were also noted.

A graduated visual analog score (VAS) 0–10 (0 = no pain and 10 = worst pain imaginable) was used to quantify pain at the following time points: induction, during the procedure, postoperatively (at 1 h after procedure), overnight, and finally the use of postoperative analgesia at any time point. Potential confounders such as surgical time and dysautonomia were noted.

Akinesia/extraocular motility (Brahma score) used for superior, inferior, medial, and lateral movement (scores: normal [3], partial movement [2], flicker [1], and no movement [0]).[5] Proptosis was assessed at the baseline and quantified by comparing with the fellow eye after the anesthetic block (non [0 mm], mild [1–2 mm], moderate [2–3 mm], and severe [>4 mm]).

Chemosis and subconjunctival hemorrhage each received a score graded as follows: (completely obscuring surgery [2], mildly obscuring surgery [1], and not obscuring surgery [0]). A gross assessment of whether the patient was able to open the eye fully (y/n) was followed by quantifying ptosis on the table (0 = non, 1 = moderate, and 2 = full). Ease of surgery (comfort of the surgeon) was on a graded VAS 0–10 (0 = no anesthetic factors contributing to increased difficulty of the procedure and 10 = anesthetic factors contributing significantly to the difficulties encountered during surgery).

Any anesthetic-related postoperative complications such as optic neuropathy and orbital cellulitis were noted. We evaluate a standard technique used in our operating theaters.


Standard descriptive statistics were used. For group differences, Chi-square test (or Fisher's exact test) was used to examine the relation between qualitative variables. For quantitative data, comparison between two groups was done using independent sample t-test or Mann–Whitney test as appropriate.


Seventy eyes of seventy patients have been used in this analysis. The mean age was 60 ± 17 years. Surgery was on the right eyes in 51% (n = 37) of patients. Surgery was vitrectomy in 100% (n = 70); this was combined with cryotherapy and gas tamponade in (n = 42), delamination and endolaser (n = 20), and internal limiting membrane/epiretinal membrane peel + gas tamponade (n = 8). Anesthesia technique used was TSTRB in 37% (n = 26), PBB in 34% (n = 24), and STB in 29% (n = 20) of patients. We did not observe any globe perforation, brain stem anesthesia, or any cardiac events.

Kallio scores were 0 in 54% (n = 38), 1 in 42% (n = 29), 2 in 4% (n = 3) of patients. Use of hyaluronidase as an adjunct 84% (n = 59) did not worsen the Kallio scale with a score of 1 in 45% (n = 27) and 2 in 5% (n = 3) (P = 0.031). Four percent (n = 3) of cases were on an anticoagulant: two on warfarin and one on rivaroxaban. In addition, three cases were on aspirin and 1 on clopidogrel. The international normalized ratio for those on warfarin was (2.3 and 2.5), within the patients aimed therapeutic window. We stopped the rivaroxaban on the day of surgery as per the standard protocol in our unit. Being on a blood thinner was not a factor for developing a subconjunctival hemorrhage (P < 0.90). All the patients on anticoagulants scored a Kallio score of 1 or less (P = 0.65). Kallio scores with various blocks are shown in [Figure 2]; the differences were statistically significant (P = 0.019).{Figure 2}

[Table 1] shows the levels of pain on the VAS for each technique. The volume of anesthesia used was a mean of 4.8 ml for TSTRB, 5.3 ml for STB, and 10.4 ml for PBB (P = 0.030). Five cases with moderate-to-severe proptosis had a volume >10 ml used (P = 0.002). [Figure 3] shows the levels of proptosis encountered with the various blocks. Intraoperative anesthesia top-up was required in 1.4% (n = 1) of patients, a PBB undergoing vitrectomy, cryotherapy, and gas. Overnight or postprocedure analgesia (a combination of paracetamol and codeine) after the procedure was required in 6% (n = 4) of cases: 2 were PBB and 2 STB; no cases from the TSTRB required analgesia after their procedure (P = 0.003). The mean VAS pain scores for the procedure was TSTRB 0.4 (± 1.4), PBB 0.4 (± 1.4), and STB 0.5 (± 1.0) (P = 0.040). Postoperative pain was TSTRB 0.2 (± 0.7), PBB 0.2 (± 0.6), and STB 1.0 (± 1.2) (P = 0.020).{Table 1}{Figure 3}

Raised IOP (28 mmHg) at day 1 was in a single case; this did not influence overnight analgesia and interpretation of our data. The mean surgical operating time was 28.8 (± 3.03) min. Longer duration of surgery was not a statistically significant factor in perioperative or postoperative pain (P = 0.500) and did not influence the need for postoperative analgesia (P = 0.700).

Ptosis of the eyelid was full in 46% (n = 32), moderate in 28% (n = 20), none in 26% (n = 18) of patients. The proportion of eyes achieving full ptosis (score of 2) with various blocks was 42% (n = 10) in PBB, 38% (n = 10) in TSTRB, and 30% (n = 6) in STB (P = 0.073).

Akinesia was 88% (n = 23) in TSTRB, 88% (n = 21) in PBB, and 80% (n = 16) in STB group. For those with some residual motility, the total Brahma was poorest in the STB 10, PBB 9, TSTRB 6 (P = 0.040) [Table 2].{Table 2}

The chemosis score was 0 as follows: TSTRB in 69% (n = 18), PBB in 91% (n = 22), and STB in 95% (n = 19). The score was 1 as follows: TSTRB in 31% (8), PBB in 8% (n = 2), and STB in 5% (1) (P = 0.028). A higher total volume of anesthesia used was not important for causing higher levels of chemosis despite the type of anesthesia block used (P = 0.574). Subconjunctival hemorrhage obscuring the view was rare, only one case in the STB group had a hemorrhage obscuring some of the surgical view scoring 1, the rest scored 0 (P = 0.702).

The volume of anesthesia used was a mean of 4.8 ml for TSTRB, 5.3 ml for STB, and 10.4 ml for PBB (P = 0.030). Five cases with moderate-to-severe proptosis had a volume >10 ml used (P = 0.002). The impact of higher levels of proptosis on ease of surgery was statistically significant (P < 0.005). Proptosis levels/technique used and ease of surgery are shown in [Table 3]. The poorest score on ease of surgery was 8 for a PBB case; the case also had severe proptosis. No statistically significant difference was found between the technique of anesthesia block used and the ease of surgery (P = 0.320) [Table 3].{Table 3}

Postoperatively, there were no cases of orbital cellulitis, optic neuropathy, or any other complications attributable to anesthesia.


A drive away from general anesthesia (GA) and sharp blocks has left some vitreoretinal surgeons relying on STB anesthesia. STB has been deemed suitable for a majority of other ophthalmic procedures such as vitreoretinal surgery, trabeculectomy, panretinal photocoagulation, optic nerve sheath fenestration, long-term postoperative pain management, and adult strabismus surgery.[6],[7],[8],[9],[10],[11] The merits and demerits of STB are procedure and surgeon dependent. The use of GA negates a majority of these issues. There is a continued trend toward LA for vitreoretinal surgery. A study in 2001 showed 82% received LA for their vitreoretinal procedure.[12]

The novel blunt cannula TSTRB described is carried out by intentionally extending a STB into the retrobulbar space, via a sub-Tenon's fenestration; it utilizes a familiar skill set. In the elderly patients, the extension sometimes occurs unintentionally, accounting for the random production dense block seen with STB at times.

In our study, being on a blood thinner did not seem to influence the level of subconjunctival hemorrhage on the Kallio scale. An important statistically significant finding was that more TSTRB cases had a subconjunctival hemorrhage (Kallio scale of 1) than the other two techniques. This could be attributable to the need to reenter the initial sub-Tenon's opening. This higher propensity for a subconjunctival hemorrhage did not seem detrimental to the anesthesia, complications, or the ease of surgery; the overall findings were of greater ease of surgery with TSTRB.

We did not find a statistically significant difference between the three techniques for pain experienced at anesthetic induction. No cases of TSTRB required analgesia after the procedure. This implies that the TSTRB analgesia is efficacious for longer periods of time, possibly due to the multilayer reservoir of anesthesia. The duration of surgery was determined to ensure that it was not a confounder for poor anesthesia due to the anesthetic wearing out; we can conclude that the findings were a fair comparison of the different anesthesia techniques and duration of surgery has not influenced our analysis.

For chemosis of the conjunctiva, the volume of anesthesia injected was not critical. It is, however, difficult to ascertain the volume of anesthesia refluxing from the block, although reflux of anesthesia typically encourages the anesthetist to reinject the LA. The volume of anesthesia actually used in the analysis might therefore be an underestimate. Only one case in the STB group had a conjunctival hemorrhage large enough to obscure the surgical view. The TSTRB also produced as much akinesia during surgery as a PBB.

An important finding in our study is that higher levels of proptosis and higher volume of anesthesia affected surgery ease significantly. It is intuitive to presume that during intraocular surgery, the volume of anesthesia behind the globe can affect the surgical ease; one such example is producing proptosis which reduces the access and maneuverability of the globe. Our study has confirmed some of the anecdotal evidence that the volume of anesthesia used is a major determinant for ease of surgery. The PBB technique seemed to require twice as much volume of anesthesia than the other two blocks and produced the most proptosis. This finding could be explained by the availability of a large potential reservoir for anesthesia to percolate into. Monitoring the effect of PBB technique is dictated by signs such as the upper eyelid dropping and reduced motility. PBB uses a sharp needle technique, reinjecting anesthesia is undesirable, and the anesthetist therefore aims to a single efficacious injection and therefore inadvertently overinjects LA. During the TSTRB, the volume seems to be less critical for adequate anesthesia as it is an enhancement of the initial sub-Tenon's insertion of anesthesia.

Sight-threatening complications with sharp needle blocks are rare studies have reported d at 2.9/10,000 with PBB and for 4.5/10,000 with PBB. A study reported 0.6/10,000 sight-threatening complications in 161,000 cases.[1] We did not observe any globe perforation, retrobulbar hemorrhage, or brain anesthesia. Postoperatively, there were no cases of orbital cellulitis, optic neuropathy, or any other complications attributable to LA. We have also not observed any of these complications with over 5000 TSTRB anesthesia blocks carried out in our unit.


We have demonstrated the TSTRB technique is both efficacious and safe. Further studies will aim to evaluate its applicability to other types of Ophthalmic surgery.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Turnbull CS. The hydrochlorate of cocaine, a judicious opinion of its merits. Med Surg Rep 1884;29:628-9.
2Eke T, Thompson JR. Serious complications of local anaesthesia for cataract surgery: A 1 year national survey in the United Kingdom. Br J Ophthalmol 2007;91:470-5.
3Sohn HJ, Moon HS, Nam DH, Paik HJ. Effect of volume used in sub-Tenon's anesthesia on efficacy and intraocular pressure in vitreoretinal surgery. Ophthalmologica 2008;222:414-21.
4Kallio H, Paloheimo M, Maunuksela EL. Haemorrhage and risk factors associated with retrobulbar/peribulbar block: A prospective study in 1383 patients. Br J Anaesth 2000;85:708-11.
5Brahma AK, Pemberton CJ, Ayeko M, Morgan LH. Single medial injection peribulbar anaesthesia using prilocaine. Anaesthesia 1994;49:1003-5.
6Kwok AK, Van Newkirk MR, Lam DS, Fan DS. Sub-Tenon's anesthesia in vitreoretinal surgery: A needleless technique. Retina 1999;19:291-6.
7Buys YM, Trope GE. Prospective study of sub-Tenon's versus retrobulbar anesthesia for inpatient and day-surgery trabeculectomy. Ophthalmology 1993;100:1585-9.
8Stevens JD, Foss AJ, Hamilton AM. No-needle one-quadrant sub-Tenon anaesthesia for panretinal photocoagulation. Eye (Lond) 1993;7 (Pt 6):768-71.
9Rizzuto PR, Spoor TC, Ramocki JM, McHenry JG. Subtenon's local anesthesia for optic nerve sheath fenestration. Am J Ophthalmol 1996;121:326-7.
10Ghosh YK, Goodall KL. Postoperative pain relief in vitreoretinal surgery with subtenon bupivacaine 0.75%. Acta Ophthalmol Scand 2005;83:119-20.
11Steele MA, Lavrich JB, Nelson LB, Koller HP. Sub-Tenon's infusion of local anesthetic for strabismus surgery. Ophthalmic Surg 1992;23:40-3.
12Newsom RS, Wainwright AC, Canning CR. Local anaesthesia for 1221 vitreoretinal procedures. Br J Ophthalmol 2001;85:225-7.