Br J Ophthalmol 96:189-192 doi:10.1136/bjo.2010.198374
  • Clinical science
  • Original article

Two-quadrant high-volume sub-Tenon's anaesthesia for vitrectomy: a randomised controlled trial

  1. David G Charteris2
  1. 1Department of Anaesthesia, Guy's and St Thomas' Hospital NHS Foundation Trust, London, UK
  2. 2Department of Anaesthesia, Moorfields Eye Hospital NHS Foundation Trust, London, UK
  1. Correspondence to Dr Vip S Gill, Department of Anaesthesia, Guy's and St Thomas' Hospital NHS Foundation Trust, Westminster Bridge Rd, London SE1 7EH, UK; drvipgill{at}
  • Accepted 13 April 2011
  • Published Online First 5 May 2011


Background Total volume using a standard single inferonasal injection for sub-Tenon's anaesthesia is limited by an increase in intraocular pressure (IOP) and commonly requires the operating surgeon to top-up the block intraoperatively. This study compares the efficacy and safety of a two-quadrant technique that allows the use of a higher volume of local anaesthetic.

Methods 54 patients undergoing vitrectomy were randomised into two groups. The control group (n=27) received a standard 5 ml single inferonasal sub-Tenon injection of a 50:50 mixture of 2% lidocaine and 0.5% bupivacaine with 150 IU hyaluronidase. The study group (n=27) received a 5 ml inferonasal and 5 ml superotemporal injection of the same mixture (10 ml total). The primary outcome measure was the number of intraoperative top-ups required. Secondary outcome measures were intraoperative and postoperative pain scores, IOP, block onset time, ocular akinesia, eyelid akinesia and chemosis.

Results 24 patients required a top-up in the control group. No patients required a top-up in the study group (p<0.001). IOP measurements were similar in both groups. Block onset was shorter, eyelid akinesia was improved and pain scores were also reduced in the study group intraoperatively and at 0–2 h, 4–6 h, 10–14 h and 20–24 h postoperatively.

Conclusions Two-quadrant sub-Tenon's anaesthesia using 10 ml of a 50:50 mixture of 2% lidocaine and 0.5% bupivacaine with 150 IU hyaluronidase seems to be more effective than a single-quadrant technique at reducing intraoperative and postoperative pain during vitrectomy.

Sub-Tenon's block has been documented as a safe and effective method to achieve local anaesthesia for vitreoretinal surgery.1–9 A perceived advantage is the avoidance of sharp needles and a theoretical reduction in their associated complications such as retrobulbar haemorrhage, globe perforation, optic nerve damage and dural puncture with subsequent brainstem anaesthesia.6 7 10–13 However, the variable quality of ocular akinesia and short block duration with subsequent need for an additional sub-Tenon's block (top-up) to be performed by the operating surgeon intraoperatively is not ideal and may be a limiting factor in its popularity for vitreoretinal surgery. Total volume and local anaesthetic mixture used varies and has an effect on block onset time, quality and duration.14 15 Local anaesthetic concentration is limited by the risk of myotoxicity to the extraocular muscles while total volume of local anaesthetic used is limited, to a degree, by the volume of the orbital cavity and an increase in intraocular pressure (IOP) as well as an increase in chemosis.16 17 A commonly used technique is a single 5 ml injection of a 50:50 mixture of lidocaine 2% and bupivacaine 0.5% via the inferonasal quadrant, which has a short onset time and reliable ocular akinesia but frequently still requires further intraoperative top-ups.7 9 14 The aim of this study was to compare the efficacy and safety of a high-volume two-quadrant sub-Tenon's technique compared with the standard low-volume single-quadrant approach for vitrectomy.


Fifty-four patients (aged 30–81 years and American Society of Anesthesiologists (ASA) grade I–III) undergoing vitreoretinal surgery (vitrectomy±laser±cryotherapy±gas exchange) for acute retinal detachment at Moorfields Eye Hospital from July 2008 to January 2009 were recruited into this prospective, randomised study. Patients allergic to local anaesthetics or hyaluronidase, having communication difficulties, requesting general anaesthesia, requiring a scleral buckle or with a scleral buckle present were excluded.

A batch of numbered sealed opaque envelopes containing each randomisation was produced using a random number generator and kept in a designated area. At recruitment, each patient was allocated a sequential envelope to be opened in the anaesthetic room only by the anaesthetist performing the block. All blocks were performed by a single anaesthetist experienced in both types of blocks. Patients were randomised to one of two groups. The control group received a 5 ml single inferonasal sub-Tenon injection of a 50:50 mixture of 2% lidocaine and 0.5% bupivacaine with 150 IU hyaluronidase.18–20 The study group received a 5 ml inferonasal injection and a 5 ml superotemporal injection of the same mixture. The principal investigator was masked to which block had been administered.

In the anaesthetic room monitoring included ECG, pulse oximetry and non-invasive arterial pressure. No sedation was given. The conjunctiva and cornea was initially anaesthetised with topical drops of proxymetacaine 0.5% followed by tetracaine 1% and sterilised with drops of 5% aqueous iodine solution. The surrounding skin was sterilised with 10% aqueous iodine solution. Non-toothed Moorfield forceps were used to grip the conjunctiva and Tenon's capsule, and then a small incision was made with Westcott scissors to expose the white sclera in the inferonasal quadrant only (control group) or inferonasal and superotemporal quadrants (study group). All injections were performed using a blunt 19G, 25 mm long curved Visitec sub-Tenon's cannula and each 5 ml injection was performed within 10 s. In keeping with the practice at our institution, no orbital pressure was applied after the block. The principal investigator then assessed the block. Ocular akinesia was assessed at 1, 3, 6, 9, 12, 15 and 18 min after the beginning of the first injection using a scoring system as described by Brahma et al in which globe movement is scored either 0 (no movement), 1 (a flicker of movement), 2 (partial movement) or 3 (full movement) in the secondary directions of gaze (abduction, adduction, elevation and depression).21 An immobile eye scored 0 and a fully mobile eye scored 12. Patients were deemed ready for surgery when the akinesia score was 4 or less. The time at which the score was 4 or less was also recorded as the block onset time. Chemosis was assessed on the number of quadrants affected. No quadrants affected scored 0 and all four quadrants affected scored 4. IOP was measured at 1, 3, 6, 9, 12, 15 and 18 min after the beginning of the first injection using a Tono-Pen (Reichert Technologies, Reichert Inc, Depew, NY, USA).22 All surgery was performed using a 20-gauge vitrectomy incision. During the procedure, the operating surgeon performed an additional sub-Tenon's block (top-up) using 3 ml of bupivacaine 0.5%, if required. Criteria for a top-up were a return of sensation or discomfort described by the patient. At the end of the procedure, type of surgery, duration of surgery and the timing of any intraoperative top-ups were recorded. The operating surgeon was also asked to score intraoperative eyelid movement from 0 to 2: 0, no movement; 1, partial function; and 2, full function.23 Pain scores were taken post-injection and postoperatively using an 11-point (0–10) numerical visual analogue scale. Patients recorded pain scores and analgesia taken at four intervals within the first 24 h after surgery: at 0–2 h, 4–6 h, 10–14 h and 20–24 h. Age, sex, ethnic origin, previous operations and postoperative complications were also recorded.

Statistical analysis

The primary outcome measure was the requirement for an intraoperative top-up. On the basis of previous studies it was assumed that the proportion of patients who require a top-up after a 5 ml injection would be 60% and that the proportion of patients who require a top-up after a 10 ml injection is 10%.7–9 From this it was calculated that 25 patients would be required per treatment group to have a study power of 95% to detect a difference at the 5% level. The numbers were then increased by 5% in each group to allow for drop-outs (two extra patients per treatment group).

The proportion of patients requiring top-ups in each group was compared using Fisher's exact test. Baseline characteristics of the treatment groups were compared to assess the adequacy of randomisation. Multiple variable logistic regression was used if any difference was detected at baseline in factors known to influence the likelihood of having a top-up. All analyses were conducted using Stata statistical software (StataCorp LP, College Station, Texas, USA).


The groups were similar with regard to patient characteristics, previous surgery, and type and duration of surgery (table 1). No patients required an additional scleral buckle and none required conversion to general anaesthesia.

Table 1

Patient characteristics, type of operation, duration of operation and type of previous surgery

The results for the study variables are shown in table 2. The main study variable (number of intraoperative top-ups) was significantly different between the two groups. Within the control group 23 patients (85.2%) required a single top-up at 50.8±11.8 min (mean±SD) and one patient (3.7%) required two top-ups. No patients required a top-up in the study group (p<0.001).

Table 2

Primary and secondary outcome measures

Block onset time was lower for the study group. Median time to an akinesia score of 4 or less was 9 min for the control group compared with 3 min for the study group. Chemosis was lower in the control group at 33.3% compared with 51.8% in the study group.

There was no significant difference in the changes in mean IOP with time between the two groups. The maximum IOP measured in the control group was 32 mm Hg and 28 mm Hg in the study group (figure 1).

Figure 1

Mean intraocular pressure (IOP) measured pre-block and at 3 minute intervals post-block. Error bars represent ±1 SD.

Lid akinesia was improved in the study group. The median (IQR) lid akinesia score was 1 (1–2) for the control group and 0 (0–0) for the study group.

Visual analogue pain scores were significantly lower for the study group compared with the control group at injection (post-block), immediately postoperatively, and at 0–2 h, 4–6 h, 10–14 h and 20–24 h postoperatively (table 2). The need for rescue analgesia within the first 24 h postoperatively was also less in the study group (table 3). In the control group two patients (7.4%) required no analgesia and 22 patients (81.5%) required codeine phosphate 30–60 mg. In the study group 12 patients (44.4%) required no analgesia and only one patient (3.7%) required codeine phosphate 30–60 mg. There were no complications in either group.

Table 3

Postoperative analgesic requirements within the first 24 h


This study shows that two-quadrant sub-Tenon's anaesthesia using 10 ml of a 50:50 mixture of lidocaine 2% and bupivacaine 0.5% with 150 IU hyaluronidase via the inferonasal and superotemporal quadrants significantly increases block duration and reduces the need for an intraoperative top-up during vitrectomy without producing any significant increase in IOP. In addition, intraoperative pain, postoperative pain and the need for rescue analgesia postoperatively were all significantly reduced compared with a single 5 ml injection of the same mixture in the inferonasal quadrant. Quality of block as assessed by the onset of ocular akinesia and perioperative eyelid movements was also significantly improved.

Our study is not without limitations. First, the study was powered using the best available evidence. However, it should be noted that all of the patients in two of these studies were given supplemental intravenous sedation.7 9 In the study by Li and colleagues 35% of the patients were given intravenous sedation.8 We did not give sedation to any of the patients in our study. Second, even though an attempt to mask the principal investigator from the type of block was made, the potential for bias must be acknowledged. There are three possible clues to indicate a high-volume two-quadrant block—a fuller looking orbit, a more proptosed eye and a second incision. Orbital fullness and proptosis may occur with a smaller volume of anaesthetic in a smaller orbit so the potential for any bias may be negated. In our practice we find that an incision after a classic inferonasal approach and both incisions after the described two-quadrant approach are concealed by the eyelids but can be identified if searched for. Finally, by the nature of the procedure and the presence of a second superotemporal incision present in the high-volume (study) group the operating surgeon could not effectively be masked to the randomisation.

Our findings for the low-volume (control) group are similar to those in previous reported studies. Kwok et al used a 4 ml single inferonasal injection of a 50:50 mixture of lidocaine 1% and bupivacaine 0.5% with 1200 IU of hyaluronidase.7 Seventy-seven per cent of patients in their study needed one or two top-ups at a mean time of 50±46 min compared with 89% in our study at a similar mean time of 50.8±11.8 min. Lai et al reported 37% of patients receiving a 5 ml single inferonasal sub-Tenon's injection of a 50:50 mixture of lidocaine 4% and bupivacaine 0.75% requiring a top-up (times of top-up not mentioned).9 Their use of a stronger anaesthetic mixture may account for the lower top-up rate. The mean intraoperative pain score in our control group was 3.4 compared with 1.17 and 2.1 in the above studies.7 9 However, all the patients in these studies were sedated with varying combinations of midazolam, alfentanil, fentanyl and propofol. None of the patients in our study were sedated. This may explain the higher top-up rate and mean intraoperative pain scores in the control group compared with previous studies. Nevertheless anaesthesia was significantly more effective in the high-volume (study) group with no one requiring a top-up compared with 89% requiring a top-up in the low-volume (control) group. Anaesthesia continued for up to 24 h postoperatively in the high-volume group, providing adequate analgesia without the need for rescue analgesia in 44.4% of patients compared with 7.4% not requiring any rescue analgesia in the low-volume group.

Cryotherapy is perceived to be the most stimulating procedure performed within our study followed by laser therapy and then gas exchange. The type of surgery performed in both groups was not significantly different. Twenty-three patients underwent cryotherapy in the high-volume group and 19 patients in the low-volume group. Patients who have had previous ocular surgery have tissue adhesions, which may limit diffusion of the local anaesthetic agent and effect. Again both groups were similar with no significant difference between the groups with regard to previous ocular surgery.

Previous studies have demonstrated sub-Tenon's anaesthesia as a safe and effective technique to achieve local anaesthesia for vitreoretinal surgery.1–9 The risk of globe compression and consequent increase in IOP associated with an injection of a volume of local anaesthetic agent into the fixed orbital space has limited the maximum volume used for sub-Tenon's anaesthesia. IOP is known to rise with increasing volume of local anaesthetic agent used for peribulbar and retrobulbar anaesthesia techniques.24 There are few studies investigating the relationship between volume and IOP after sub-Tenon's anaesthesia.17 25 26 There are no comparative data observing the effect of injecting 10 ml of local anaesthetic into the sub-Tenon space on IOP in the current literature. An important new finding in this study is that injection of 10 ml of local anaesthetic agent into the sub-Tenon space via the inferonasal and superotemporal quadrants did not cause any significant rise in IOP. In fact we observed a small but significant gradual reduction in IOP up to 18 min after injection. Alwitry et al used 5 ml of injectate via the inferonasal quadrant and also observed a small but insignificant rise in IOP followed by a significant reduction in IOP.17 Sohn et al, however, found significant rises in IOP with 3 ml, 5 ml and 7 ml of injectate injected via the inferonasal quadrant alone.26 These differing responses in IOP might be due to the differing ethnic patient populations being investigated. Sohn et al suggested that their findings might in part be due to the smaller orbital volume and tighter orbital septum found more commonly among the Asian population.26 The reduction in IOP observed in our study and that of Alwitry et al may be a direct pharmacological effect of the local anaesthetic agent on ocular blood flow, or secondary to a reduction in aqueous production due to an effect on the ciliary ganglion, or possibly because of a reduction in extraocular muscle tone.17

Block quality as assessed by onset time and eyelid akinesia was also significantly improved in the two-quadrant high-volume group. A median reduction in block onset of 6 min may not be clinically important, but continual intraoperative eyelid movements are distracting for the surgeon, interfere with surgery and may affect clinical outcome.

The incidence of chemosis increases as the volume of local anaesthetic injected into the sub-Tenon space increases.4 13 14 Our findings are consistent with this and the incidence of 52% in the high-volume group is within the range of 25% to 60% reported by previous studies using a lower volume of local anaesthetic agent.6 This is not usually a surgical problem—chemosis interferes less with vitreoretinal surgery compared with anterior segment surgery and usually resolves after the application of digital pressure. This may, however, be an issue where small gauge trans-conjunctival vitreoretinal surgery is performed but in our opinion is unlikely to influence the choice of the vitrectomy incision size.

In conclusion, our study has shown that 10 ml of local anaesthetic agent may be injected via the inferonasal and superotemporal quadrants into the sub-Tenon space without significantly increasing IOP. A 10 ml 50:50 mixture of lidocaine 2% and bupivacaine 0.5% with 150 IU hyaluronidase provides significantly better block quality, perioperative anaesthesia and postoperative analgesia for vitrectomy compared with a standard single-quadrant low-volume (5 ml) technique using the same mixture. Further studies are required to resolve whether this is due to an increased dose of local anaesthetic agent, increased spread around the orbit and into the retrobulbar space secondary to an increase in volume of local anaesthetic or due to increased spread associated with the anatomy of a two-quadrant approach.


  • Funding This study was funded by Moorfields Eye Hospital NHS Foundation Trust.

  • Competing interests None declared.

  • Patient consent Obtained.

  • Ethics approval This study was conducted with the approval of the Moorfields & Whittington Hospitals Local Research Ethics Committee (NHS REC Ref: 08/H0721/33).

  • Provenance and peer review Not commissioned; externally peer reviewed.


Relevant Article