Endophthalmitis is a sight-threatening eye infection that can complicate any type of intraocular surgery including vitreoretinal surgery. The incidence, risk factors, management, microbiology, and prognosis of endophthalmitis following vitreoretinal surgery have been well-described, which demonstrated that this complication is rare but has a poor prognosis.1,2
It is a common practice to administer either a subconjunctival or intracameral antibiotic at the end of cataract surgery to reduce the risk of postoperative endophthalmitis.3 In vitreoretinal surgery, the anterior segment is often not breached; thus, rather than the antibiotic being delivered into the anterior chamber, it is often administered by subconjunctival injection. However, such practice is debatable, and its benefit is largely unproven. Subconjunctival antibiotic injections, such as cefuroxime (a second-generation cephalosporin) has also been described to be associated with postoperative pain.4
Local anesthesia for vitreoretinal surgery is routinely achieved by either sub-Tenon's or retrobulbar regional anesthesia.5 Vitreoretinal surgery at the Toronto Western Hospital (Toronto, Canada) typically is performed with retrobulbar anesthesia involving the transcutaneous injection of anesthetic solution, consisting of 2% lidocaine and 0.5% marcaine, into the retrobulbar space using a 25-gauge needle. At the end of surgery, subconjunctival cefazolin (Ancef; GlaxoSmithKline, Brentfort, United Kingdom) and hydrocortisone sodium succinate (Solu-Cortef; Pfizer, New York City, NY) are given.
Although a previous study explored the reduction of pain through the use of buffered lidocaine injected in the subconjunctival space before cefuroxime for cataract surgery,4 the utility of pain reduction using 2% lidocaine with cefazolin before subconjunctival injection has not yet been evaluated in a formal randomized control trial. Cefazolin-lidocaine combination may serve as an effective strategy to minimize pain associated with prophylactic subconjunctival antibiotic delivery after retinal surgery.
Patients and Methods
Patients requiring retinal surgery at Toronto Western Hospital (Toronto, Canada) were included in this study. Ethics approval was obtained from the University Health Network Research Ethics Board, and the trial was registered in a national registry ( www.clinicaltrial.gov, NCT02324166; accessed December 2014). A sample size calculation was performed prior to study commencement. A difference of two points on a 0-to-10 pain intensity numerical rating scale was determined to be significant.6 A standard deviation of 2.4 was used from a previous study with a power of 80%, and an alpha-error of 5%.5 Noting that the pain scale reflects an ordinal variable, using the worst-case scenario for the Pitman Asymptomatic Relative Efficiency (ARE), the sample size for a t test (as above) was divided by 0.864. From these numbers, 27 subjects were calculated per group, for a total “n” value of 54. Only one eye per patient was included in the study sample. Patients were randomized using simple randomization through a simple number generator. Concealment of the randomization list was accomplished via distance randomization. Informed consent was obtained from all patients prior to vitreoretinal surgery in addition to separate informed consent to participate in the study. The study was performed following the tenets of the Declaration of Helsinki and the Good Clinical Practices guidelines.
Patients 18 years of age or older who were able to give informed consent and who were undergoing retinal surgery for the first time were included in this study. Other inclusion criteria included patients requiring retrobulbar anesthesia for vitreoretinal surgery, and that patients were able to provide consent to the study. Exclusion criteria included if the patient was cognitively incapable of performing the study, or if they were unable to fully understand the study requirements and provide informed consent. Patients were recruited from two retinal surgical practices (RGD and WCL) at Toronto Western Hospital, Department of Ophthalmology, Toronto, Canada.
On the day of surgery, study subjects were randomly assigned to receive either a 0.5 mL cefazolin-balanced salt solution consisting of 0.3 mL of cefazolin and 0.2 mL of balanced salt solution or 0.5mL cefazolin-lidocaine solution consisting of 0.3 mL of cefazolin and 0.2 mL of 2% lidocaine. Through this methodology, both groups received the same dose of cefazolin and same overall injection volume. Both groups received the injection in the subconjunctival space at the end of their retinal surgery while the operating ophthalmologist stated the following standardized line: “I will be administering the antibiotic now to prevent infection.” This served to mask participants on which group they were assigned, as a marked time point.
In the recovery room within 30 minutes after retinal surgery, patients completed the numerical pain scale relating to their operative experience with a research assistant or clinical fellow. The scale consisted of a linear line divided into 10 equal intervals, with the leftmost one marked “0,” indicating no pain, and the rightmost one marked “10,” representing the worst pain imaginable. If the patient was unable to see the scale or read the accompanying text, the scale was described, and the independent research assistant obtained a verbal response. Patients were retrospectively asked to rate their pain using the scale at four points: a) at the start of surgery b) during the middle of surgery c) at the end of surgery immediately before the doctor gave the antibiotic, and d) immediately postoperatively after the doctor injected the antibiotic. Basic demographic information such as gender, race, age, and diagnosis, as well as type of procedure performed was also recorded. Procedures were categorized into either phacoemulsification with intraocular lens (IOL) insertion and 23-gauge pars plana vitrectomy (PPV) with additional procedures or 23-gauge PPV with additional procedures, where additional procedures included laser, air fluid exchange, membrane peel, or cryotherapy. The anesthesia record for each operation was reviewed to retrieve each patient's American Society of Anesthesiologists (ASA) physical status classification; the number of anesthetic drugs (ie, midazolam, fentanyl, or propofol) given in the first, second, and last third of the operation, respectively; the patient's vitals (ie, blood pressure and heart rate) at the beginning, middle, and end of the operation; and the patient's past non-ophthalmologic surgical history.
Statistical analysis was performed using SPSS (Version 22.0; IBM; Armonk, NY). Ordinal variables, including comparisons between intergroup pain rating scores and the ASA, were reported with the median as the measure of central tendency and analyzed through a Mann-Whitney U test. Continuous variables such as heart rate, systolic blood pressure, or diastolic blood pressure were reported as the mean as the measure of central tendency and analyzed through student's t test. Categorical variables such as having a reported past non-ophthalmologic surgical history versus no past non-ophthalmologic surgical history, were compared through a Chi-squared test. Statistical significance, with the exception of the Chi-squared test, were two-tailed and performed at the alpha level of 0.05.
A total of 54 patients were recruited in this study; 11 were recruited from RGD's surgical practice, whereas 43 were recruited from WCL's surgical practice (20.4% vs. 79.6%, respectively). Of these 54 patients, half were randomized to the cefazolin with lidocaine group and half were randomized to cefazolin only group. Of the study population, 44.4% of patients were male and the mean age was 60.1 years ± 13.5 years. The top four indications for surgery were epiretinal membrane (ERM) (21.1%), cataract with ERM (15.8%), cataract and vitreous hemorrhage (12.3%), and vitreous hemorrhage (12.3%). Of the procedures performed, 46.3% had phacoemulsification with IOL insertion and PPV with additional procedures, whereas 53.7% had PPV with additional procedures (Table 1).
Summary of Demographics
When comparing surgical characteristics between the two study groups, there was no statistically significant difference between ASA score, duration of surgery, number of intravenous drugs administered, or vitals at the preoperative, intraoperative, and postoperative time points (Table 2). Patients who were randomized to the cefazolin-only group were more likely to have had a past nonophthalmologic surgical history (P = .05) (Table 2). However, neither study group was more likely to have one type of surgical procedure over the other (P = 0.79) (Table 2).
Study Group Operative Characteristics
The majority of patients in both study groups reported no pain (0 of 10) or mild pain (1 of 10) at the start of the operation, middle of surgery, before or after the cefazolin injection, or postoperatively. However, at each of the time points examined, there was no statistically significant difference (P > .05) in pain scores between the two groups (Table 3). Since it was not the intention of the study, postoperative complications were not assessed.
Pain Scores Between Study Groups
Subconjunctival delivery of antibiotics is a common method of prophylaxis against endophthalmitis after retinal surgery,3 and a mixture of lidocaine in the subconjunctival space before administration of cefuroxime has been reported to reduce pain associated with the procedure.4 This is the first prospective, randomized, single-masked clinical study examining whether subconjunctival cefazolin mixed with 2% lidocaine will minimize pain levels compared to cefazolin alone.
This is also the first study to assess the pain associated with subconjunctival cefazolin injection in the setting of retinal surgery. The majority of patients in the group receiving cefazolin alone reported a pain level of “0” after the subconjunctival injection and postoperatively. This may suggest that there is little discomfort associated with the subconjunctival delivery of cefazolin, or local anesthetic provided sufficient analgesia.
Adverse reactions to a lidocaine-cefazolin mixture delivered subconjunctively are rare. There has been one case report in the English literature describing a lidocaine hypersensitivity reaction to a subconjunctival lidocaine, cefazolin, and gentamicin mixture.7
There are number of study limitations that should be noted. Firstly, pain scores were assessed through subjective patient recall at the end of the surgery. With the administration of sedating agents such as midazolam, fentanyl, or propofol, recall and reduction of painful stimulus can be affected during awake sedation.8 Although there was no statistically significant difference between the number of intravenous sedating agents during the first third, second third, or last third of the surgery (Table 2) between the two groups, sedating agents may theoretically can impair the accuracy of pain reporting for study subjects. Patients can have variable responses to anesthesia, which can affect their level of recall and subsequently survey results. Secondly, surgeons were not masked to the group allocations, although the assignment of the treatment group is randomly generated by the computer. This may have subtlety affected the way the statement, “I will be administering the antibiotic now to prevent infection,” was delivered and interpreted by the patients, and potentially could have confounded results.
Anecdotally, some patients experience significant discomfort near the end of retinal surgery at the time of subconjunctival cefazolin injection. This study demonstrated, though, that the majority of patients did not feel any pain postoperatively in both groups; no statistical differences were found between the two groups in terms of pain. The two theories behind potential discomfort caused by subconjunctival injection of cefazolin at the end of retinal surgery observed by the authors may be due to stretching of the conjunctival tissue or the acidity of the injection.9 Future studies can consider pre-administration of topical tetracaine or subconjunctival lidocaine before subconjunctival delivery of cefazolin. Alternatively, the addition of bicarbonate mixed with cefazolin can be considered as a strategy to minimize patient discomfort during subconjunctival injection of cefazolin by neutralizing the acidity of the injection.
Although this study and our center utilizes a retrobulbar block, some surgeons may choose to utilize a peribulbar block with subconjunctival lidocaine or general anesthesia.10 In either method, lidocaine or marcaine at the end of the case may be beneficial for pain control.
In conclusion, this prospective study of 54 patients undergoing retinal surgery randomized to either subconjunctival cefazolin with 2% lidocaine or cefazolin alone reported no statistically significant difference in pain scores between groups. In PPV with or without phacoemulsification and IOL insertion, subconjunctival cefazolin mixed with 2% lidocaine confers no analgesic benefit.
- Park JC, Ramasamy B, Shaw S, Ling RHL, Prasad S. A prospective and nationwide study investigating endophthalmitis following pars plana vitrectomy: Clinical presentation, microbiology, management and outcome. Br J Ophthalmol. 2014;98(8):1080–1086. doi:10.1136/bjophthalmol-2013-304486 [CrossRef]
- Park JC, Ramasamy B, Shaw S, Prasad S, Ling RHL. A prospective and nationwide study investigating endophthalmitis following pars plana vitrectomy: Incidence and risk factors. Br J Ophthalmol. 2014;98(4):529–533. doi:10.1136/bjophthalmol-2013-304485 [CrossRef]
- Gower EW, Lindsley K, Nanji AA, Leyngold I, McDonnell PJ. Perioperative antibiotics for prevention of acute endophthalmitis after cataract surgery. Cochrane Database Syst Rev. 2013;(7):CD006364.
- Kayarkar VV, Dinakaran S. Topical anesthesia for phacoemulsification and painless subconjunctival antibiotic injection. J Cataract Refract Surg. 2001;27(2):198–200. doi:10.1016/S0886-3350(00)00564-2 [CrossRef]
- Lai MM, Lai JC, Lee WH, et al. Comparison of retrobulbar and sub-Tenon's capsule injection of local anesthetic in vitreoretinal surgery. Ophthalmology. 2005;112(4):574–579. doi:10.1016/j.ophtha.2004.10.043 [CrossRef]
- Younger J, McCue R, Mackey S. Pain outcomes: A brief review of instruments and techniques. Curr Pain Headache Rep. 2009;13(1):39–43. doi:10.1007/s11916-009-0009-x [CrossRef]
- Levy J, Lifshitz T. Lidocaine hypersensitivity after subconjunctival injection. Can J Ophthalmol. 2006;41(2):204–206. doi:10.1139/I06-010 [CrossRef]
- Wilson TD, McNeil DW, Kyle BN, Weaver BD, Graves RW. Effects of conscious sedation on patient recall of anxiety and pain after oral surgery. Oral Surg Oral Med Oral Pathol Oral Radiol. 2014;117(3):277–282. doi:10.1016/j.oooo.2013.11.489 [CrossRef]
- U.S. Pharmacopeia. USP Monographs: Cefazolin Ophthalmic Solution. http://www.pharmacopeia.cn/v29240/usp29nf24s0_m13980.html. Published 2014. Accessed May 19, 2016.
- Palte H. Ophthalmic regional blocks: management, challenges, and solutions. Local Reg Anesth. 2015;8:57–14. doi:10.2147/LRA.S64806 [CrossRef]
Summary of Demographics
|Demographic Features||Study Group|
| Male, % (n)||44.4% (24)|
| Female, % (n)||55.6% (30)|
|Age (Years), Mean ± SD, Range (n)||60.1 ± 13.5, 28–87 (54)|
|Indication for Surgery, % (n)|
| Cataract and epiretinal membrane||16.7% (9)|
| Cataract and macular hole||7.4% (4)|
| Cataract and proliferative diabetic retinopathy||3.7% (2)|
| Cataract and retinal detachment with vitreous hemorrhage||1.9% (1)|
| Cataract and vitreous hemorrhage||13.0% (7)|
| Epiretinal membrane||22.2% (12)|
| Macular hole||7.4% (4)|
| Retinal detachment||11.1% (6)|
| Retinal detachment with vitreous hemorrhage||1.9% (1)|
| Vitreous hemorrhage||13.0% (7)|
| Vitreous hemorrhage with proliferative diabetic retinopathy||1.9% (1)|
|Procedures Performed, % (n)|
| Phacoemulsification with intraocular lens insertion and PPV with additional procedures1||46.3% (25)|
| PPV with additional procedures1||53.7% (29)|
Study Group Operative Characteristics
|Operative Characteristics||Cefazolin Without Lidocaine||Cefazolin With Lidocaine||P Value|
|ASA Score: Median, Range (n)||2, 1–4 (27)||2, 1–4 (27)||.611|
|Duration of Surgery (Minutes), Mean ± SD, Range (n)||59.4 ± 20.9, 30–105 (27)||66.1 ± 17.4, 30–100 (27)||.212|
|Past Surgical History|
| No past surgical history||25.9% (7)||51.9% (14)||.053|
| Reported past surgical history||74.1% (20)||48.1% (13)|
|Surgical Procedures Per Study Group|
| Phacoemulsification with IOL and PPV with additional procedures||44.4%||48.1%||.793|
| PPV with additional procedures||55.6%||51.9%|
|Number of IV Sedating Agents Administered±|
| First third of surgery: median, range (n)||4, 3–9 (27)||4, 3–8 (27)||.921|
| Second third of surgery: median, range (n)||0, 0–5 (27)||0, 0–3 (27)||.561|
| Last third of surgery: median, range (n)||0, 0–2 (27)||0, 0–3 (27)||.671|
| Systolic blood pressure: mean ± SD, range (n)||130.6 ± 16.7, 99–160 (26)||126.9 ± 16.4, 100–160 (27)||.422|
| Diastolic blood pressure: mean ± SD, range (n)||76.15 ± 10.0, 49–93 (26)||75.9 ± 10.3, 60–92 (27)||.932|
| Heart Rate: mean ± SD, range (n)||73.1 ± 10.3, 50–90 (26)||70.4 ± 9.4, 55–88 (26)||.322|
| Systolic blood pressure: mean ± SD, range (n)||133.8 ± 25.2, 95–190 (27)||129.7 ± 18.0, 100–170 (27)||.502|
| Diastolic blood pressure: mean ± SD, range (n)||71.3 ± 11.5, 55–90 (27)||68.4 ± 9.6, 55–90 (27)||.322|
| Heart Rate: mean ± SD, range (n)||64.8 ± 9.0, 50–85 (27)||64.1 ± 13.0, 40–95 (27)||.842|
| Systolic blood pressure: mean ± SD, range (n)||140.3 ± 25.3, 100–190 (27)||132.3 ± 17.0, 105–170 (27)||.182|
| Diastolic blood pressure: mean ± SD, range (n)||73.3 ± 11.4, 50–95 (27)||71.9 ± 9.3, 60–95 (27)||.622|
| Heart Rate: mean ± SD, range (n)||67.1 ± 11.4, 50–95 (27)||64.9 ± 12.6, 42–95 (27)||.512|
Pain Scores Between Study Groups
|Time Point in Surgery||Cefazolin Without Lidocaine, Mean Pain Score, Median Pain Score, Range (n)||Cefazolin With Lidocaine, Mean Pain Score, Median Pain Score, Range (n)||P Value|
|Start of Surgery||0.44, 0, 0–8 (27)||0.72, 0, 0–8 (27)||.08|
|Middle of Surgery||1.33, 0, 0–6 (27)||2.04, 1, 0–7 (27)||.34|
|Before Cefazolin Injection||1.46, 0, 0–8 (26)||1.48 0, 0–6 (27)||.57|
|After Cefazolin Injection||2.37, 1, 0–9 (27)||2.24, 1, 0–7 (27)||.88|
|Postoperative||1.52, 0, 0–8 (25)||1.30, 0, 0–5 (27)||.93|