Ocular Immunology and Inflammation

ISSN: 0927-3948 (Print) 1744-5078 (Online) Journal homepage: [Link]

Safety and Outcome of Microincision Vitreous

Surgery in Uveitis

Reema Bansal MS, Amod Gupta MS, Vishali Gupta MS, Samyak Mulkutkar
MS, Mohit Dogra MS, Deeksha Katoch MS, Mangat R. Dogra MS, Kusum
Sharma MD, Mini P. Singh MD, Aman Sharma MD, Shivali Kamal PhD & Surya
P. Sharma MSc

To cite this article: Reema Bansal MS, Amod Gupta MS, Vishali Gupta MS, Samyak Mulkutkar
MS, Mohit Dogra MS, Deeksha Katoch MS, Mangat R. Dogra MS, Kusum Sharma MD, Mini P.
Singh MD, Aman Sharma MD, Shivali Kamal PhD & Surya P. Sharma MSc (2016): Safety and
Outcome of Microincision Vitreous Surgery in Uveitis, Ocular Immunology and Inflammation,
DOI: 10.3109/09273948.2016.1165259

To link to this article: [Link]

Published online: 18 May 2016.

Submit your article to this journal

Article views: 20

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

[Link]

Download by: [University of Nebraska, Lincoln] Date: 05 June 2016, At: 07:56
 Ocular Immunology and Inflammation, 2016; 00(00): 1–10
© Taylor & Francis Group, LLC
ISSN: 0927-3948 print / 1744-5078 online
DOI: 10.3109/09273948.2016.1165259

ORIGINAL ARTICLE

Safety and Outcome of Microincision Vitreous

Surgery in Uveitis
Reema Bansal, MS1, Amod Gupta, MS1, Vishali Gupta, MS1, Samyak Mulkutkar, MS1, Mohit Dogra,
1 1 1 2 3
MS , Deeksha Katoch, MS , Mangat R. Dogra, MS , Kusum Sharma, MD , Mini P. Singh, MD , Aman
Sharma, MD4, Shivali Kamal, PhD1, and Surya P. Sharma, MSc1

1
Department of Ophthalmology, Advanced Eye Centre, Post Graduate Institute of Medical Education and
Research, Chandigarh, India, 2Departments of Medical Microbiology, Post Graduate Institute of Medical
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

Education and Research, Chandigarh, India, 3Virology, Post Graduate Institute of Medical Education and
Research, Chandigarh, India, and 4Medicine, Post Graduate Institute of Medical Education and Research,
Chandigarh, India

ABSTRACT
Purpose: To report the outcome of microincision vitreous surgery (MIVS) in uveitis.
Methods: In total, 103 patients (106 eyes) underwent diagnostic MIVS between March 2012 and April 2015.
Postoperative evaluation included vitreous haze grading from clinical/electronic records, best-corrected visual
acuity (BCVA), and complications.
Results: Mean age was 36.8 ± 13.9 years (range: 8–80 years). Mean follow-up after MIVS was 12.2 ± 7.2 months
(median 12 months). Mean vitreous haze grading was 2.39 ± 0.98 (preoperatively), 0.36 ± 0.73 postoperatively
(1 week), and 0.02 ± 0.2 at 1 month (p < 0.001). Mean BCVA was 1.5 ± 1.0 logMAR preoperatively and 0.72 ±
0.68 logMAR at 1 month (p = 0.000). Postoperative complications included cataract (14.6%), rise in intraocular
pressure (13.2%), vitreous hemorrhage (4.7%), hypotony (3.2%), retinal detachment (2.8%), epiretinal mem-
brane (2.8%), and worsening of inflammation (0.9%).
Conclusions: MIVS is safe and may have a therapeutic role in uveitis.
Keywords: Microincisional vitrectomy, uveitis, vitritis

Diamond and Kaplan (1979) first demonstrated evaluated the safety and tolerance of microincision
improved visual outcomes following pars plana (23G or 25G) PPV in uveitis eyes in a large cohort.
vitrectomy (PPV) and lensectomy in patients with
uveitis, and postulated that removing the vitreous gel
alone was therapeutic in uveitis.1 Subsequently, MATERIALS AND METHODS
besides its therapeutic effect, the conventional
20-gauge PPV was also found useful in obtaining We conducted a subgroup analysis of safety and tol-
intraocular samples in cases that posed a diagnostic erance of 23G/25G MIVS in uveitis patients who were
challenge.2,3 However, the fear of surgically-induced prospectively enrolled from March 2012 to April 2015
exacerbation of intraocular inflammation limited its at the Advanced Eye Centre, Post Graduate Institute of
role to either cases of poor visibility of the retina or Medical Education and Research, Chandigarh, India.
high suspicion of intraocular malignancy.4,5 In recent Institute Ethics Committee approval was obtained
years, the advent of microincision vitreous surgery prior to the study. The study adhered to the tenets of
(MIVS) appears to have changed this outlook.6–9 We the Declaration of Helsinki.

Received 11 December 2015; revised 25 February 2016; accepted 7 March 2016; published online 19 May 2016
Correspondence: Amod Gupta, Department of Ophthalmology, Advanced Eye Centre, Post Graduate Institute of Medical Education and
Research, Chandigarh 160012, India. E-mail: dramodgupta@[Link]
Color versions of one or more of the figures in the article can be found online at [Link]/ioii.

1
 2 R. Bansal et al.

In our clinic, all patients with uveitis undergo a 6. Severe uveitis precluding retinal evaluation
detailed history and clinical evaluation. Additionally, 7. Pathologies co-existing with uveitis requiring
those with intermediate, posterior, or panuveitis definitive surgical intervention (epiretinal mem-
undergo ancillary testing including fundus photogra- brane, vitreous hemorrhage, retinal detachment).
phy, fundus autofluorescence, fluorescein angiography
(FA), and optical coherence tomography (OCT). Patients with the following were excluded:
Indocyanine green angiography (ICGA) or B-scan
ultrasound are performed as and when required. 1. Post-traumatic endophthalmitis
Empirical therapy is initiated in patients with a specific 2. Acute postoperative endophthalmitis
uveitis entity diagnosed on the basis of a typical clin- 3. Anterior uveitis.
ical presentation, or when corroborated by targeted
laboratory tests, such as tuberculin skin test (TST), Data was retrieved from the clinical chart records of
chest X-ray, or CT chest, Treponema pallidum hemag- the patients for: age; gender; laterality of uveitis;
glutination and serologic tests for HIV. Invasive MIVS type (anatomic and morphologic) of uveitis; labora-
is performed by an experienced surgeon as and when tory test results; suspected etiology of uveitis; type
indicated (Figure 1). and duration of treatment (antibiotics, corticoster-
A written informed consent was obtained from all oids, or immunosuppressive agents); best-corrected
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

patients before surgery. The patients who completed at visual acuity (BCVA) and intraocular pressure (IOP)
least 3 months of follow-up after MIVS and who met before and after surgery; final diagnosis after mole-
the following inclusion criteria were enrolled in the cular/cytologic test results of vitreous fluid analy-
study: sis; and follow-up. Information relevant to surgical
details included indications for MIVS, additional
vitreoretinal procedure, and complications (early
1. Intermediate, posterior, or panuveitis of and late) if any. We analyzed the digital images of
unknown or suspected origin, where conven- all eyes for vitreous haze grading before and after
tional methods (clinical assessment and labora- PPV.10
tory investigations) failed to determine the Patients requiring vitreous surgery for sequelae (and
diagnosis were quiescent at the time of surgery) were not adminis-
2. Unresponsive to conventional antibiotic/corti- tered oral corticosteroids. All patients underwent a stan-
costeroid/immunosuppressive therapy dard 23-G or 25-G MIVS through a wide-viewing
3. Strong suspicion of intraocular malignancy system. The vitreous samples were collected and trans-
4. A potentially sight-threatening acute uveitis with ferred to the respective laboratories in cold storage boxes.
a negative non-invasive laboratory and ancillary Vitreous base dissection, fluid-air exchange, scleral buck-
investigations ling, and internal tamponade (gas/silicon oil) were per-
5. Atypical presentation of uveitis formed as deemed necessary. Closure of the sclerotomy

FIGURE 1. Flow chart of diagnostic vitrectomy for uveitis in our clinic.

Ocular Immunology and Inflammation

 Microincisional Vitrectomy in Uveitis 3

sites by suturing was performed in selected cases at the TABLE 1. Demographic and baseline clinical details of 103
surgeon’s discretion. Patients receiving oral corticoster- patients (106 eyes) with uveitis subjected to pars plana
vitrectomy.
oids for active uveitis at the time of undergoing MIVS
were tapered off this therapy during their postoperative Number of patients (eyes) 103 (106)
follow-up, depending upon the clinical response. Males:Females 69:34
Postoperative improvement of inflammation was Age (years)
assessed in terms of BCVA and vitreous haze from Mean ± SD 36.8 ± 13.9
Range 8–80
clinical as well as electronic records. Worsening of Unilateral PPV:Bilateral PPV 100:3
inflammation was assessed by the appearance of new Uveitis
cells in the anterior chamber/vitreous cavity, or a Intermediate 20
decline in BCVA, and increasing vitreous haze. Posterior 68
The primary outcomes were an improvement in Panuveitis 18
Follow up after initial presentation
inflammation after MIVS and postoperative complica- (months)
tions. The secondary outcome measures included Mean ± SD 23.04 ± 36.1
improvement in BCVA, the diagnostic yield of MIVS, Median 13
and duration of corticosteroids/immunosuppressive Range 3 months to 22
therapy. years
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

Follow up after PPV (months)

Mean ± SD 12.2 ± 7.2
Median 12
Statistical Analysis Range 3–34 months

SD, standard deviation; PPV, pars plana vitrectomy.

The BCVA on Snellen’s chart was converted into a
logarithm of minimum angle of resolution (logMAR)
for statistical analysis. Descriptive statistics included
mean, median, and range. Student’s paired t-test was therapeutic MIVS for vitreous hemorrhage (14); retinal
applied to quantitative variables before and after PPV. detachment (12); and epiretinal membrane (1).
All the statistical analyses were done using SPSS soft- The mean preoperative BCVA (in 106 eyes) was
ware version 17 (SPSS, Inc, Chicago, IL) for Windows 1.5 ± 1.0 logMAR (median 1.3, range: 0–3). Pars
at 5% level of significance. A value of p<0.05 was plana vitrectomy was combined with cataract surgery
considered statistically significant. in five eyes (two pars plana lensectomy, and three
phacoemulsification with intraocular lens implanta-
tion) and posterior capsulectomy in one eye. Sixteen
RESULTS eyes underwent internal tamponade [10 gas tampo-
nade (SF6/C3F8), and six silicon oil tamponade]. On
Of 118 patients (122 eyes) with uveitis undergoing MIVS, the first postoperative day, the mean BCVA was 1.14 ±
15 patients (16 eyes) had an insufficient follow-up 0.85 logMAR [as compared with preoperative 1.4 ±
(<3 months) after MIVS. We analyzed 103 patients (106 0.99 logMAR in 92 eyes (after excluding eight eyes
eyes), which included 69 males and 34 females. The mean with SF6/C3F8 tamponade; two eyes with cataract
age was 36.8 ± 13.9 years (range: 8–80 years). Three surgery combined with gas tamponade; four eyes
patients underwent MIVS in both eyes (Table 1). The with cataract surgery/posterior capsulectomy); p =
mean follow-up of these patients in our clinic (from initial 0.005, median = 1, range: 0–3] (Table 3). At 1 week,
visit to final visit) was 23.04 ± 36.1 months (median 13 the mean BCVA was 0.89 ± 0.75 logMAR [as com-
months, range: 3 months to 22 years), and the mean pared with preoperative 1.41 ± 0.99 logMAR in 97
follow-up after PPV was 12.2 ± 7.2 months (median 12 eyes (after excluding three gas-filled eyes with SF6/
months, range: 3–34 months). C3F8 tamponade; two eyes with cataract surgery com-
A total of 27 (25.5%) eyes had chronic, persistent bined with gas tamponade; four eyes with cataract
inflammation of long duration, with poor response surgery/posterior capsulectomy); p = 0.000, median =
to conventional empirical therapy (Table 2). Of 0.6, range: 0–3]. At 1 month, the mean BCVA was
these, 12 were receiving immunosuppressive ther- 0.69 ± 0.66 logMAR in 100 eyes [as compared with
apy (azathioprine). Acute, sight-threatening uveitis preoperative 1.44 ± 0.99 logMAR (after excluding six
necessitated PPV in 22 (20.7%) eyes. Thirteen (12.3%) eyes with cataract surgery/posterior capsulectomy);
eyes had uveitis of unknown or suspected origin, where p = 0.000, median = 0.5, range: 0–3] (Table 3). Hence,
conventional methods (clinical assessment and labora- the mean BCVA achieved at any visit (1 day, 1 week,
tory investigations) failed to determine the etiology. or 1 month) following PPV showed significant
Atypical uveitis and severe uveitis precluding retinal improvement from the mean preoperative BCVA.
evaluation were seen in seven (6.6%) eyes each. Three A total of 21 eyes had follow-up <6 months, and 85
(2.8%) eyes had a strong suspicion of intraocular malig- eyes had a minimum follow-up of 6 months (29 eyes >6
nancy. Additionally, 27 (25.5%) eyes underwent months but <12 months, and 56 eyes >12 months). Six

© 2016 Taylor & Francis Group, LLC

 4 R. Bansal et al.

TABLE 2. Indications for pars plana vitrectomy (PPV) in patients with uveitis.

Eyes

Indications n (%)

1 Co-existing pathologies requiring 27 25.5

definitive surgical intervention
(epiretinal membrane, vitreous hemorrhage, retinal detachment)
Vitreous hemorrhage 14 13.2
Retinal detachment 12 11.3
Epiretinal membrane 1 1.0
2 Unresponsive to conventional antibiotic/corticosteroid/immunosuppressive therapy 27 25.5
(or long-standing recurrent inflammation)
3 A potentially sight-threatening acute uveitis to initiate a specific therapy 22 20.7
(or to confirm suspected diagnosis in some cases)
4 Uveitis of unknown or suspected origin, where conventional methods 13 12.3
(clinical assessment and laboratory investigations) fail to determine the diagnosis
5 Severe uveitis precluding retinal evaluation 7 6.6
6 Atypical presentation 7 6.6
7 Strong suspicion of intraocular malignancy 3 2.8
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

eyes were excluded from analysis of BCVA and vitr- The mean preoperative prednisolone dose (in 86
eous haze due to simultaneous cataract surgery with patients, 87 eyes) was 28.22 ± 21.20 mg/day (median =
MIVS (one of these had follow-up <6 months; 1 had 30 mg/day, range: 5–80 mg/day) and the mean post-
follow-up between 6 and 12 months; and four had operative dose of corticosteroids was 5.52 ± 6.68 mg/
follow-up >12 months). Of 80 eyes (after excluding day (range: 0–30 mg/day, median = 5 mg/day)
five eyes that underwent cataract surgery with MIVS) (p<0.001). These could be stopped altogether in 43
that completed at least 6 months of follow-up after patients (43 eyes) during their course of follow-up,
MIVS, the mean pre-and postoperative BCVA at 6 and could be tapered to low-dose maintenance ther-
months was 1.37 ± 0.983 logMAR units, and 0.61 ± apy (≤7.5 mg/day) in 16 patients (16 eyes). The dose of
0.661 logMAR units, respectively (p<0.001). Of 52 eyes 7.5 mg/day was taken as the cut-off, as we consider
that completed at least 12 months of follow-up after this as the acceptable maintenance dose of oral
MIVS (after excluding four eyes that underwent catar- prednisone.11 Steroid therapy with/without steroid-
act surgery with MIVS), the mean pre-and postopera- sparing immunosuppressive therapy (azathioprine in
tive BCVA at 12 months was 1.27 ± 0.96 logMAR units, 12 patients) was continued in 27 patients (28 eyes). In
and 0.61 ± 0.72 logMAR units, respectively (p<0.001) patients continuing oral steroids at the final visit, the
(Table 3). mean postoperative dose (in 44 eyes) was 10.8 ± 5.4
There were no differences in the visual outcomes of mg/day (range: 5–30 mg/day, median = 10 mg/day).
MIVS between the acute and chronic uveitis eyes Overall, the mean postoperative dose of corticosteroids
(data not shown). Excluding six eyes (with cataract in 86 patients (87 eyes) who received preoperative
surgery/posterior capsulectomy) from vitreous haze corticosteroids was 5.6 ± 6.7 mg/day (range: 0–30
grading analysis, the mean preoperative vitreous mg/day, median = 5 mg/day). The difference between
haze grading in the remaining 100 eyes was 2.39 ± pre- and postoperative corticosteroids was statistically
0.98 (median = 2, range: 1–5). Postoperatively, it was significant (p<0.001).
0.36 ± 0.73 (median = 0, range: 0–3) at 1 week and Overall, oral corticosteroids could be stopped or
0.02 ± 0.2 (median = 0, range: 0–2) at 1 month maintained on low dose in 59/87 (67.8%) eyes follow-
(p<0.001) (Table 4). At 6 months (after excluding one ing MIVS (Table 5).
more eye with significant cataract (awaiting cataract Preoperatively, 85 eyes were clinically suspected to
surgery) post-MIVS within 6 months), the mean pre- have a specific etiology. Following MIVS, a definitive
and postoperative (6 months) vitreous haze score in etiology could be obtained in 57 eyes. Table 6 shows
79 eyes was 2.35 ± 1.01 and 1.0 ± 0, respectively. the diagnostic yield of PPV in 106 eyes. In addition to
Similarly, after excluding two more eyes with signifi- the therapeutic benefits, MIVS helped in reaching a
cant cataract post-MIVS within 12 months, the mean correct diagnosis and subsequent change in treatment
pre- and postoperative (12 months) vitreous haze strategy. Five of the 21 with presumed idiopathic uvei-
score in 50 eyes was 2.2 ± 0.95 and 1.0 ± 0, respec- tis were positive for TB by polymerase chain reaction
tively (Table 4). and hence received anti-tuberculosis treatment in addi-
In total, 86 patients (87 eyes) were receiving oral tion to systemic steroids. Of 85 presumed infective
corticosteroids for active uveitis at the time of MIVS. uveitis cases, 33 were re-labelled as idiopathic and

Ocular Immunology and Inflammation

 Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

TABLE 3. Preoperative and postoperative best-corrected visual acuity (logMAR) at various follow-up visits.

Follow-up visits 1 day (92 eyes) 1 week (97 eyes) 1 month (100 eyes) 6 months (80 eyes) 12 months (52 eyes)

BCVA (logMAR) Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative

© 2016 Taylor & Francis Group, LLC

Mean ± SD 1.4 ± 0.99 1.14 ± 0.85 1.41 ± 0.99 0.89 ± 0.75 1.44 ± 0.99 0.69 ± 0.66 1.37 ± 0.98 0.61 ± 0.66 1.27 ± 0.96 0.61 ± 0.72
Range 0–3 0–3 0–3 0–3 0–3 0–3 0–3 0–3 0–3 0–3
Median 1 1 1 1 0.5 0.6 1 0.5 1 0.4
p value 0.005 0.000 0.000 0.000 0.000

BCVA, best-corrected visual acuity; SD, standard deviation.

TABLE 4. Preoperative and postoperative vitreous haze grading at various follow-up visits.

Follow-up visits 1 week (100 eyes) 1 month (100 eyes) 6 months (79 eyes) 12 months (50 eyes)

Vitreous haze grading Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative Preoperative Postoperative

Mean ± SD 2.39 ± 0.98 0.36 ± 0.73 2.39 ± 0.98 0.02 ± 0.2 2.35 ± 1.01 1.0 ± 0 2.2 ± 0.95 1.0 ± 0
Range 1–5 0–3 1–5 0–2 0–3 – 0–3 –
Median 2 0 2 0 1 – 1 –
p value 0.000 0.000 0.000 0.000
Microincisional Vitrectomy in Uveitis 5
 6 R. Bansal et al.

TABLE 5. Overall distribution of eyes of patients (undergoing pars plana vitrectomy) requiring
oral corticosteroids and their status of corticosteroid therapy at the final visit following surgery.

Preoperative oral corticosteroids Postoperative oral corticosteroids p valuea

Administered (n = 86 patients, 87 eyes) Stopped (43 eyes, 49.4%) 0.274

≤7.5 mg/day (32 eyes, 36.78%) ≤7.5 mg/day (16 eyes, 18.4%)
>7.5 mg/day (55 eyes, 63.22%) >7.5 mg/day (28 eyes, 32.2%)
Not administered (n = 17 patients, 19 eyes) None
a 2
χ -test.

TABLE 6. Preoperative and postoperative diagnosis (106 eyes).

Postoperative diagnosis

Preoperative diagnosis Idiopathic TB Viral PVRL Fungal

Idiopathic (n = 21) 16 5 – – –
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

TB (n = 67) 22 44 1 – –
Viral (n = 8) 5 – 3 – –
Toxoplasma (n = 2) 2 – – – –
Toxocara (n = 2) 2 – – – –
PVRL (n = 3) 1 – – 2 –
Fungal (n = 2) 1 – – – 1
Syphilis (n = 1) – 1 – – –
Total eyes (n = 106) 49 50 4 2 1

TB, tuberculosis; PVRL, primary vitreoretinal lymphoma.

required only steroids. The cause of infection was for ERM peeling, two for vitreous hemorrhage, and one
changed in two cases (TB to viral in one, and syphilis for postoperative hypotony with suprachoroidal
to TB in one). hemorrhage).
At the time of undergoing MIVS, four eyes were
pseudophakic, one aphakic, and five eyes had significant
cataract. Of the other 96 phakic eyes, postoperative cat- CASE EXAMPLES
aract developed in 14 (14.6%) eyes (Table 7). Suturing of
sclerotomy ports was done in 12 eyes at the time of Case 1
surgery. Hypotony occurred in three (3.2%) of the 94
eyes that were not sutured at the time of vitrectomy. Of A 25-year-old woman presented with decreased vision
these, one eye developed suprachoroidal hemorrhage (counting fingers) in the right eye for 12 years. She had
that resolved after an additional surgery that required significant vitritis and extensive choroiditis in the right
its drainage. Three eyes (2.8%) developed retinal detach- eye (Figure 2a). The left eye was normal. Her TST was
ment postoperatively, requiring a repeat vitreous sur- 20 × 20 mm, and chest CT was normal. Following
gery (preoperatively, one eye had retinal detachment MIVS, her BCVA improved to 6/18 on the 1st post-
with acute retinal necrosis, one had vasculitic vitreous operative day, and media cleared significantly
hemorrhage, and one had choroidal granuloma). Five (Figure 2b). The vitreous fluid analysis was positive
eyes developed vitreous hemorrhage postoperatively, for Mycobacterium tuberculosis by PCR. She received
of which three resolved spontaneously, and two under- anti-tubercular therapy (ATT) with oral corticoster-
went vitreous lavage. Epiretinal membrane developed in oids. At 14 months follow-up, her BCVA was 6/18 in
three eyes (2.8%) following MIVS. Only one eye (0.9%) the right eye (Figure 2c).
showed worsening of inflammation, which resolved by
hiking the oral corticosteroids and adding immunosup-
pressive drugs. A transient rise in IOP was seen in 14 Case 2
(13.2%) eyes (13 in the late postoperative period and one
following silicon oil tamponade), which were managed A 23-year-old male had bilateral, healed, multifocal
with anti-glaucoma medications. None of the eyes devel- choroiditis with active subretinal granulomas in the
oped postoperative wound leakage or endophthalmitis. left eye, showing hyperfluorescence and leakage on
Overall, a repeat MIVS was performed in eight eyes FA (Figure 3a,b). The BCVA was 6/18 in the right
(three for retinal detachment after diagnostic PPV, two eye and hand movements in the left eye. His TST

Ocular Immunology and Inflammation

 Microincisional Vitrectomy in Uveitis 7

TABLE 7. Complications following pars plana vitrectomy in uveitis eyes.

Eyes 3 days to 1
Within 3 days of month of 1–3 months of 3 months after
Complications n (%) MIVS MIVS MIVS MIVS

Cataract 14/96 14.6 – – 4 10

Raised intraocular pressure 14/106 13.2 2 1 5 6
(>21 mmHg)
Vitreous hemorrhage 5 4.7 5 – – –
Hypotony 3/94 3.2 3 – – –
Retinal detachment 3 2.8 – 1 1 1
Epiretinal membrane 3 2.8 – – – –
Worsening of inflammation 1 0.9 1 – – –

was positive, and the chest X-ray was normal. cytokines) and infectious antigens may be responsible
Following MIVS in the left eye, his BCVA improved for eliminating inflammation.
to 6/60 at 1 week, with the resolution of vitritis Besides fundus evaluation, angiographic evidence
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

(Figure 3c). Not only the media clarity improved, but of decreased inflammation has been reported by way
there was also a significant reduction in the activity of of reduction in cystoid macular edema after
the granuloma, as seen by minimal leakage on FA vitrectomy.1 Our second case demonstrates clinical as
(Figure 3d). The TB-PCR was positive. He received well as angiographic evidence of decreased inflamma-
ATT with oral corticosteroids. The granulomas healed tion in the form of decreased leakage from choroidal
at 3 months with BCVA of 6/60 in the left eye granulomas.
(Figure 3e). Development or progression of cataract is well
known after PPV, with rates of 51%, 39%, and 33%
reported previously.3,4,6 In our study, 14.6% of phakic
eyes showed cataract progression (or development)
DISCUSSION over a mean follow-up of 12.2 ± 7.7 months after
MIVS. A higher rate in a previous study may be due
The earlier uncertainties about the role of PPV in uvei- to a longer follow-up period (mean 17.2 ± 11.4
tis have been largely overcome by a series of small months).4
reports of its usefulness in these eyes.2 More recent Unusual postoperative IOP responses, abnormal
papers have reported a decrease in inflammatory bleeding (intra- or postoperative) and exaggerated
activity and a decrease in uveitis flares after vitrect- inflammatory response have been major concerns
omy, along with an increase in visual acuity. related to an intraocular surgery in uveitis eyes.15
Moreover, recent small series of MIVS have high- Following 25-G vitrectomy, transient hypotony in the
lighted the beneficial role and safety of PPV in eyes early postoperative period is common.16 Though a
with uveitis.6–8 When compared with 20-G vitreous majority of such cases recover spontaneously, the
surgery for other posterior segment conditions (such potential risks associated with hypotony cannot be
as idiopathic epiretinal membranes, rhegmatogenous ignored. Androudi et al. reported a single case (6.6%)
retinal detachments or proliferative diabetic retinopa- of extreme hypotony in a series of 15 cases of 25-G
thy), MIVS has been found to be less invasive and is MIVS that responded to an intensive steroid therapy.6
believed to minimize postoperative inflammation.12–14 In a larger series of sarcoid uveitis patients undergoing
We observed a significant reduction of vitreous MIVS, rubeotic glaucoma developed in one eye, and
haze in our patients, as early as the next postoperative none developed hypotony.7 Kitiratschky et al. had one
day. A significant reduction of posterior segment patient each developing ocular hypertony and hypot-
inflammation has been noted in patients with sarcoi- ony after 20-G PPV.3 In our series, three (3.2%) of 94
dosis from 1 week after MIVS.7 The authors used the eyes that were not sutured during vitrectomy, devel-
same score of vitreous inflammation as we used.10 oped hypotony. Two eyes recovered spontaneously,
They also reported a significant decrease in anterior and one with the hemorrhagic choroidal detachment
segment inflammation after 1 month in eyes with was successfully treated with a repeat vitreous surgery
MIVS, and after 12 months in eyes with MIVS with for suprachoroidal drainage, with a final BCVA of 6/9
phacoemulsification. The mechanism by which vitrect- at 33 months of follow-up. In 20-G PPV in uveitis,
omy improves inflammation is not known precisely. overall 54% eyes developed complications (cataract
Removal of vitreous that harbors the inflammatory 51%, retinal detachment 2%, epiretinal membrane 7%,
cells/mediators (including immune complexes and vitreous hemorrhage 2%, and hypotony 2%.).4

© 2016 Taylor & Francis Group, LLC

 8 R. Bansal et al.
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

FIGURE 2. (a) Fundus photograph of the right eye of a 25-year-old woman showing significant vitritis and extensive choroiditis. (b,c)
Fundus photograph of the same eye, 1 day after PPV (b), showing improvement in media haze, and at 14 months of follow-up (c).

Postoperative bleeding was observed in five (4.7%) statistically significant improvement in visual acuity,
eyes in our series, while Androudi et al. and starting one week postoperatively.7 We found a signifi-
Kitiratschky et al. reported none (in 15 cases and 70 cant improvement in the mean BCVA at all visits fol-
cases, respectively).3,6 Takayama et al. reported one lowing MIVS (1 day, 1 week, or 1 month) as compared
case (out of 24 eyes) of recurrence of vitreous hemor- with the mean preoperative BCVA, showing progres-
rhage after vitrectomy.7 Among other complications, sive improvement with time (p<0.05). We believe that
proliferative vitreoretinopathy has been reported beneficial visual effects of MIVS are derived from a
uncommonly.6,7 Retinal detachment occurred after variety of factors, such as surgical removal of inflamed
PPV in 7% of patients in a previous study3 and vitreous, corticosteroids, and combined cataract
three (2.8%) eyes in our series. Preoperative acute removal. It is possible that PPV may facilitate the
retinal necrosis was present in three and one of the ingress of helpful vascular mediators to reach the vitr-
eyes, respectively, which was associated with an eous cavity.
increased postoperative complication rate.4 We did A decreased use of immunosuppressive therapy has
not see postoperative endophthalmitis or corneal been associated with PPV.1,17 In our series, preoperative
decompensation in any of the patients. Worsening of systemic corticosteroids could be stopped (or tapered to
intraocular inflammation as a direct adverse effect of low dose ≤7.5 mg/day) in 67.8% eyes. Oahalou et al.
vitreous removal or surgical stress was not reported were able to stop preoperative immunosuppressive
in previous studies.4,6,7 It was seen in only one eye in therapy in 44% of patients after PPV.4 In a small, pro-
our series (0.9%). spective study of patients with recalcitrant intermediate
Postoperatively, an epiretinal membrane developed uveitis, 82% (9/11) of eyes treated with PPV showed a
in three (2.8%) eyes in our study, while 20-G PPV in long-term resolution of inflammation, as compared
uveitis was associated with 7% eyes developing epir- with 43% (3/7) of eyes treated with immunomodula-
etinal membrane.4 Proliferative vitreoretinopathy as a tory therapy, the rest subsequently requiring PPV.17
postoperative complication has been reported in a sin- Also, PPV proved superior to immunomodulatory ther-
gle eye after 25-G MIVS.7 apy in terms of improvement in visual acuity and vitr-
Improvement in visual acuity and inflammatory eous cell reduction.17
activity after PPV in uveitis has been reported in several Our results have limitations of a short follow-up.
studies.1,4,6–8 Androudi et al. reported visual benefits in Baseline FA or OCT for central macular thickness
a majority of their patients.6 Takayama et al. observed a could not be used as an outcome measure as it

Ocular Immunology and Inflammation

 Microincisional Vitrectomy in Uveitis 9
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

FIGURE 3. (a,b) Fundus photograph (a) of the left eye of a 23-year-old male with healed multifocal choroiditis and active subretinal
granulomas, and fluorescein angiography (b) showing hyperfluorescence and leakage. (c,d) Fundus photograph (c) of the same eye 1
week after PPV, showing resolution of vitritis and healing of granuloma, and fluorescein angiography (d) showing minimal leakage
from the granuloma. (e) Fundus photograph at 3 months of follow-up after PPV.

could not be performed in some eyes due to dense DECLARATION OF INTEREST

vitreous inflammation or significant preoperative
cataract. The authors report no conflicts of interest. The authors
While microincision PPV has previously been advo- alone are responsible for the content and writing of the
cated safe and efficacious in selected uveitis cases,6 our paper.
study reflects its usefulness as a therapeutic measure in a
wide variety of complicated uveitis cases, such as cataract
with uveitis, dense vitritis, vitreous hemorrhage, trac- FUNDING
tional, or rhegmatogenous retinal detachment, and sub-
retinal biopsy, etc. Besides reducing the need for long- This study was supported by the Department of
term systemic corticosteroids/immunosuppressive ther- Biotechnology, New Delhi, India.
apy, an early MIVS in patients with posterior uveitis with
significant vitreous inflammation may avoid the need for
additional sub-tenon or intravitreal steroid injections that REFERENCES
pose potential adverse effects in the eye. Extensive fundus
scarring resulting from long-standing, uncontrolled chor- 1. Diamond JG, Kaplan HJ. Uveitis: effect of vitrectomy com-
oidal inflammation (as seen in Figure 2) may also be bined with lensectomy. Ophthalmology. 1979;86:1320–1329.
2. Becker M, Davis J. Vitrectomy in the treatment of uveitis.
avoided by timely vitrectomy, yielding better visual Am J Ophthalmol. 2005;140:1096–1105.
results. We found MIVS to be safe and efficacious, and 3. Kitiratschky VBD, Deuter C, Beck R, et al. Relationship
well tolerated by the majority of eyes. between suspected reasons of intraocular inflammation

© 2016 Taylor & Francis Group, LLC

 10 R. Bansal et al.

and the results of diagnostic vitrectomy: an observational intermediate and posterior uveitis. Ophthalmology.
study. Ocul Immunol Inflamm. 2015;23:59–66. 1985;92:467–471.
4. Oahalou A, Schellekens PAWJF, De Groot-Mijnes JD, 11. Esterberg E, Acharya NR. Corticosteroid-sparing therapy:
et al. Diagnostic pars plana vitrectomy and aqueous ana- practice patterns among uveitis specialists. J Ophthalmic
lyses in patients with uveitis of unknown cause. Retina. Inflamm Infect. 2012;2:21–28.
2014;34:108–114. 12. Bamonte G, Mura M, Stevie Tan H. Hypotony after 25-G
5. Foster CS, Opremcak EM. Therapeutic surgery: cornea, iris, vitrectomy. Am J Ophthalmol. 2011;151:156–160.
cataract, glaucoma, vitreous, retinal. In: Foster CS, Vitale A, 13. Rizzo S, Genovesi-Ebert F, Murri S, et al. 25-gauge, suture-
eds. Diagnosis and Treatment of Uveitis. Philadelphia, PA: WB less vitrectomy and standard 20-gauge pars plana vitrect-
Saunders; 2002:222–235. omy in idiopathic epiretinal membrane surgery: a
6. Androudi S, Praidou A, Symeonidis C, et al. Safety and comparative pilot study. Graefes Arch Clin Exp Ophthalmol.
efficacy of small incision, sutureless pars plana vitrectomy 2006;244:472–479.
for patients with posterior segment complications second- 14. Kobayashi S, Sato S, Inoue M, et al. Comparison of 20- and
ary to uveitis. Acta Ophthalmol. 2012;e409–e410. 25-gauge vitrectomy for primary repair of rhegmatogenous
7. Takayama K, Tanaka A, Shibata M, et al. Evaluation of retinal detachment. Ophthalmic Surg Lasers Imaging.
microincision vitrectomy surgery using a wide-viewing sys- 2010;41:550–554.
tem for complications with ocular sarcoidosis. Medicine. 15. Kiryu J, Kita M, Tanabe T, et al. Pars plana vitrectomy for
2015;94:e559. epiretinal membrane associated with sarcoidosis. Jpn J
8. Soheilian M, Ramezani A, Soheilian R. 25-gauge vitrectomy for Ophthalmol. 2003;47:479–483.
complicated chronic endogenous/autoimmune uveitis: predic- 16. Sato T, Emi K, Bando H, et al. Faster recovery after 25-gauge
Downloaded by [University of Nebraska, Lincoln] at 07:56 05 June 2016

tors of outcomes. Ocul Immunol Inflamm. 2013;21:93–101. microincision vitrectomy surgery than after 20-gauge vitrect-
9. Bansal R, Sharma K, Gupta A, et al. Detection of omy in patients with proliferative diabetic retinopathy. Clin
Mycobacterium tuberculosis genome in the vitreous fluid Ophthalmol. 2012;6:1925–1930.
of eyes with multifocal serpiginoid choroiditis. 17. Quinones K, Choi JY, Yilmaz T, et al. Pars plana vitrectomy
Ophthalmology. 2015;122:840–850. versus immunomodulatory therapy for intermediate uvei-
10. Nussenblatt RB, Palestine AG, Chan CC, et al. tis: a prospective, randomized pilot study. Ocul Immunol
Standardization of vitreal inflammatory activity in Inflamm. 2010;18:411–417.

Ocular Immunology and Inflammation