Abstract
In-office vitreous biopsy is currently performed with a 25-gauge needle or less frequently with a specialized in-office surgical system. This article demonstrates in-office vitreous biopsy with a standard vitreous cutter, using syringes to actuate the cutter. A 79-year-old woman presented six days after intravitreal bevacizumab with endophthalmitis. After subconjunctival anesthesia, a valved 27-gauge trocar was inserted through the pars plana. Two syringes were connected to a pneumatic 27-gauge Alcon vitrectomy handpiece and manually actuated by an assistant while the physician aspirated with a third syringe to obtain the vitreous biopsy. Intravitreal vancomycin and ceftazidime were injected. A total of 0.5 cc of fluid was collected without complications. Manual actuated vitrectomy reliably collects sufficient vitreous samples for diagnostic evaluation and may be safer and more effective than needle biopsy.
Vitreous biopsy is important for the diagnosis and management of endophthalmitis, malignancy, and uveitis.1,2 However, between 0% to 35% of vitreous taps using a needle are dry, which limits the ability to diagnose the causative organism and susceptibility to antibiotics.3,4 To improve diagnostic yield, portable vitreous cutting consoles and instrumentation for office-based vitreous biopsy and vitrectomy have been developed.5 However, these systems require the purchase and maintenance of special equipment for a relatively rare situation. Recently, a trocar cannula system has been used with a 25-gauge needle for vitreous biopsy.6 This provides a single access point for both removal of fluid and insertion of intravitreal medication. The current technique describes an office-based vitreous biopsy technique using a 27-gauge trocar/cannula with a 27-gauge vitreous cutter manually actuated by an assistant. Standard off-the-shelf surgical instruments are used without the need to purchase an additional vitreoretinal surgical machine.
THE CASE
A 79-year-old woman presented with right eye pain and decreased vision for three days. Past medical history was remarkable for diabetic macular edema of both eyes, and she had received 1.25 mg of intravitreal bevacizumab to both eyes six days prior. Separate lots of bevacizumab, speculums, and supplies were used for each eye. Vision at that time was 20/60 in both eyes. On examination, visual acuity was hand motions in the right eye and 20/60 in the left eye. Anterior segment examination of the right eye was remarkable for beefy raised conjunctiva from 7:00 to 11:00, 4+ cell, 2+ flare, 1.5 mm hypopyon, and a posterior chamber intraocular lens. Dilated fundoscopic examination demonstrated grade 5 vitritis with no view to the posterior pole. B-scan of the left eye showed dense vitritis without retinal detachment. Left eye examination was unchanged from prior.
After discussion with the patient, vitreous biopsy using the technique reported was performed with a standard Alcon 27-gauge vitreous cutter. A total of 0.5 cc of vitreous was cut, aspirated, and sent to the laboratory for gram stain, bacterial, and fungal culture. Intravitreal vancomycin (1 mg/0.1 mL) and ceftazidime (2.25 mg/0.1 mL) were provided, and the cannula was removed. The patient was started on topical moxifloxacin, prednisolone acetate, and oral moxifloxacin. Cultures grew Staphylococcus epidermidis resistant to azithromycin, erythromycin, and penicillin. Pain improved, and at post-injection month two, visual acuity was 20/200. 25-gauge pars plana vitrectomy with intravitreal vancomycin and ceftazidime was performed for vitreous opacities. At post-injection month six, the patient’s vision was 20/40.
SURGICAL TECHNIQUE
The following supplies were placed on a sterile field: a 27-gauge Alcon Vitrectomy Pak, one Viscous Fluid Control Pak (VFC), one 1-cc syringe for vitreous collection, two 10-cc syringes, one sterile syringe cap, a sterile jeweler’s forceps, and sterile scissors. Intravitreal ceftazidime and vancomycin, two 30-gauge 5/8-inch needles, betadine, subconjunctival 2% lidocaine in a 1cc syringe, standard eye rinse, and cotton tips were prepared.
The 27-gauge vitreous cutter was addressed first (Figure 1). The Luer lock from the middle aspiration line connected to a 1-cc syringe for aspiration (1). The distal portion of the line was removed. The distal ends of the drive lines were cut with scissors. The barbed Luer fitting from the infusion line (2) and from the VFC (3) were removed and attached to the drive lines and were ensured to be securely fit. The sterile 10-cc syringes (4) were brought to the 5-cc mark and attached to the drive lines. Actuation was checked by first pushing one of the 10-cc syringe to move the guillotine to the end of travel, and then releasing pressure on that syringe and actuating the second 10-cc syringe to retract the guillotine. The vitreous cutter could be heard to actuate as the guillotine reached the end of travel.
Figure 1.

Components for vitreous biopsy. A pneumatic Alcon 27-gauge vitreous cutter is attached to two 10-cc syringes from the drive lines and a single 1-cc syringe from the aspiration port. Adapters for the drive lines are taken from the Viscous Fluid Control (VFC) Pak and from the 27-gauge Vitrectomy Pak.
The intravitreal antibiotics were prepared with a sterile 30-gauge 5/8-inch needle. Subconjunctival lidocaine 2% was placed in the inferotemporal quadrant of the affected eye followed by topical betadine solution. A speculum was placed, and the surgeon put on sterile gloves. The inferotemporal quadrant was marked 3.0 mm from the limbus for pseudophakic patients and 4.0 mm from the limbus for phakic patients with the back of the trocar. The valved cannula was inserted with a beveled incision using the 27-gauge trocar/cannula.
The vitreous cutter was provided sterile to the surgeon with the 3-cc aspiration syringe, which was used by the surgeon’s left hand. The vitreous cutter was inserted into the trocar, and the assistant was instructed to begin cutting. At this time, the assistant alternated actuating the two syringes to actuate the cutter. Cut rate was approximately 60 cuts per minute, and vitreous was aspirated. After aspiration of an adequate sample, aspiration was stopped. Vitreous cutting was then stopped to prevent vitreous incarceration in the cutter. The cutter was removed from the eye, residual vitreous in the cutter and line were aspirated into the syringe, and the syringe was removed from the drive line, capped with a sterile cap, and sent to microbiology.
The intravitreal vancomycin (1 mg/0.1 mL) and ceftazidime (2.25 mg/0.1 mL) were injected through the cannula into the eye. The cannula was removed with the sterile jeweler’s forceps with a cotton tip to ensure closure of the wound. The eye was washed out with eye rinse in standard fashion.
DISCUSSION
In this article, we describe a surgical technique for in-office vitreous biopsy using a trocar/cannula system with a manually actuated 27-gauge vitreous cutter. We have used this procedure on three patients to date. In our experience, use of a vitreous cutter has reduced the rate of dry taps, improved biopsy volume, and likely reduced peripheral traction and tears of the retina.
In addition to use with endophthalmitis, this technique can be used for collection of in-office vitreous samples for diagnosis of malignancy or noninfectious uveitis. This is particularly desirable for patients who cannot tolerate anesthesia or are poor surgical candidates.
Although this technique is described for the Alcon vitreous cutter with dual pneumatic drive, we believe this technique can be used with other vitreous cutters and are actively exploring techniques with less assembly. Prior to use, we observe cutter actuation using a microscope to ensure adequate closure, as incomplete closure could cause vitreous incarceration in the cutter and peripheral traction.
Acknowledgments:
The authors thank Teja Chemudupati, Department of Ophthalmology, Stanford University, for assistance with the video and images. This retrospective study was approved by the Institutional Review Board and adhered to the tenets set forth in the Declaration of Helsinki. All participants provided written informed consent.
Financial disclosures:
CMTD reports grants from Research to Prevent Blindness and the National Institutes of Health (NIH; P30-EY026877, K12 EY033745); VBM reports grants from Research to Prevent Blindness and the NIH (P30-EY026877, R01EY024665, R01EY025225, R01EY024698, R21AG050437); and SRS reports grants from Research to Prevent Blindness and the NIH (P30-EY026877).
REFERENCES
- 1.Patnaik G, Annamalai R, Biswas J. Intraocular biopsy in uveitis. Indian J Ophthalmol. 2020;68(9):1838–1843. 10.4103/ijo.IJO_1325_20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Group EVS; Endophthalmitis Vitrectomy Study Group. Results of the Endophthalmitis Vitrectomy Study. A randomized trial of immediate vitrectomy and of intravenous antibiotics for the treatment of postoperative bacterial endophthalmitis. Arch Ophthalmol. 1995;113(12):1479–1496. 10.1001/archopht.1995.01100120009001 [DOI] [PubMed] [Google Scholar]
- 3.Pfahler SM, Brandford AN, Glaser BM. A prospective study of in-office diagnostic vitreous sampling in patients with vitreoretinal pathology. Retina. 2009;29(7):1032–1035. 10.1097/IAE.0b013e3181a2c1eb [DOI] [PubMed] [Google Scholar]
- 4.Giovinazzo JV, Jansen M, Iacob C, Rosen RB, Deobhakta A. Obtaining vitreous samples in acute endophthalmitis: comparing the rate of dry taps using a needle vs trocar cannula. Invest Ophthalmol Vis Sci. 2018;59(9):3673. [Google Scholar]
- 5.Josephberg RG, Hussain A. Vitreous biopsy for endophthalmitis, uveitis, or cytology. In: Operative techniques in vitreoretinal surgery. 2022:104–112. [Google Scholar]
- 6.Chan A, Leng T, Moshfeghi DM. Cannula-based 25-gauge vitreous tap and injection: a new surgical technique. Retina. 2012;32(5):1021–1022. 10.1097/IAE.0b013e318248e6ba [DOI] [PubMed] [Google Scholar]
