Peer-Reviewed: Intraocular Pressure Changes After Removal of Sub-Tenon’s Triamcinolone Acetonide Depot


Intraocular Pressure Changes After Removal of Sub-Tenon's Triamcinolone Acetonide Depot


Steroid-induced ocular hypertension (SIOH) is a complication often seen in patients treated with periocular and intravitreal steroid injections. Of the periocular corticosteroids, the depot preparations present a particular risk.1 Severe, intractable intraocular pressure (IOP) elevation is a potential sequela of anterior sub-Tenon's triamcinolone (STT) acetonide injection that must be considered by all clinicians when determining whether to perform the procedure.2 It has been reported that this procedure can result in irreversible ocular hypertension.3-6 However, Herschler2 and Kalina et al.7 demonstrated that this process may be treated by surgically excising the triamcinolone depot.

Many factors contribute to the predisposition of SIOH, such as family history and history of primary open angle glaucoma. Diabetes, high myopia and connective tissue diseases can cause higher rates of steroid responsiveness compared to the general population.3 Younger age, female gender, and higher baseline IOP are other risk factors for SIOH.8,9 It is estimated that one in every three people is considered a potential steroid responder — some to acute levels.10 It has been reported that approximately 5% of the adult population are considered high responders (increases in IOP >15 mm Hg); as many as 3% of steroid responders may experience persistent and irreversible IOP elevation.6

The exact mechanism of SIOH continues to be debated, but it is generally accepted that the facility of aqueous outflow is decreased. Some have noted an increase in aqueous inflow, while others have reported an effect on both inflow and outflow.4 Steroids cause cellular, biochemical and molecular changes in the trabecular meshwork (TM).13 They increase deposition of extracellular material in the trabecular beams, fingerprint-like deposits in the uveal meshwork and fibrillar deposits in the juxtacanalicular tissue of the TM.3 This extracellular material is acid mucopolysaccharide, or glycosaminoglycan (GAG), resulting in reduced facility of aqueous outflow. It was proposed by Francois that steroids stabilize lysosomes in fibroblasts of the meshwork which would prevent the release of hyaluronidase, leading to the accumulation of polymerized GAGs in the TM.15

It appears that corticosteroids have an affinity for TM cells, especially the nuclei, as shown by observations suggesting a high concentration of steroid-specific receptors.5 It is postulated that steroid treatment may upregulate the MYOC gene (which is responsible for the production of the protein myocilin) in susceptible TM cells.14 Excess extracellular myocilin has the potential to bind to the aqueous outflow pathways in the TM and increase outflow resistance. It can also bind to ciliary muscle, having an impact on its ability to control IOP. These structures help regulate IOP.

Potency, duration, route of administration and dose of the steroid being used contribute to the probability of developing SIOH. A general order of increasing effect on IOP is as follows: systemic steroids < topical steroids < periocular steroid injections < intravitreal steroid injections. Patients who have been treated with steroids may see an elevation in IOP within days, weeks, months or even years after initiating therapy.5 One study showed the median onset of increased IOP to be as short as three weeks after STT.11

The question of duration of action of STT depot therapy is not without controversy. Although Kalina et al.7 determined that there was pharmacologically active triamcinolone up to 13 months following injection, others have suggested a duration of action as short as three months. A prospective study done by Jonas showed that, after an intravitreal injection of triamcinolone acetonide, measurable concentrations could be detected in aqueous humor samples up to 1.5 years after the application.12

Because periocular steroids are given to control inflammation and edema, clinicians are faced with the possibility of causing an exacerbation of the underlying condition being treated. This might occur when attempting to minimize the IOP-related side effects by considering periocular steroid depot removal. The purpose of this study was to determine whether the removal of periocular steroid depots (anterior sub-Tenon's triamcinolone acetonide in this case series) would indeed lead to reversibility of the SIOH. However, a secondary objective was to determine whether or not a loss of therapeutic benefit as manifested by visual acuity was associated with any potential IOP reversibility.


An exploratory/descriptive report was performed, based on a consecutive longitudinal retrospective case series of 25 eyes from 22 patients who underwent sub-Tenon's triamcinolone removal (STTR) due to SIOH secondary to STT in a single private practice setting in Pittsburgh, PA, between November 2004 and June 2007. All patients carried a diagnosis of cystoid macular edema (CME) from various etiologies. Patients were not excluded based on underlying diagnosis. No patients carried a prior diagnosis of open-, narrow- or closed-angle glaucoma. Demographic, IOP (primary endpoint), and Snellen visual acuity (VA, secondary endpoint) data, as well as number of topical IOP-lowering agents (IOPLAs) were obtained at various time intervals for all subjects, including time of STT (baseline), one week after STTR (post-STTR one week), four weeks after STTR (post-STTR four weeks), and 12 weeks after STTR (post-STTR 12 weeks). Best VA (post-STT best), maximum IOP (post-STT max), and maximum IOPLA following STT were also documented. Mean results were calculated for each of these data points. IOP was calculated using Haag-Streit (Mason, OH) Goldmann applanation tonometry.

Data for the baseline and post-STT max examinations were complete. However, the follow-up examinations included substantial missing data: for IOP, 19% (14/75) of the follow-up datapoints were missing in 32% (8/25) of the eyes; for VA and IOPLA, 13% (10/75) of the follow-up datapoints were missing in 24% (6/25) of the eyes. Given these levels of missing data, no attempts were made to statistically estimate the missing values. The dependent variables (IOP, VA, IOPLA) were found to be strongly non-normal in distribution. This factor, together with our limited sample size and the levels of missing data, argued for the use of a simple nonparametric procedure to test for statistical differences among clinical time points. A series of pairwise comparisons were carried out using a Wilcoxin signed-rank test in order to account for the paired nature of the data. Statistically significant P-values emerging from the Wilcoxin tests of each dependent variable were sequentially adjusted for multiple comparisons using the Holm-Sidak23 procedure. Primary statistical analyses were carried out on eyes, rather than individuals, in order to maximize our statistical power and to fully utilize the limited available data. All of the analyses in this report were subsequently carried out a second time utilizing only a single randomly selected (when appropriate) eye from each patient. No significant modifications to the findings of the primary analysis emerged from this secondary analysis.

The procedure of STT was as follows: Each patient had undergone STT for the treatment of CME. The procedure was done in an outpatient office setting. Informed consent was obtained. The patient was placed in a supine position at approximately 45° in the examination chair. A quarter-inch ribbon of 1% lidocaine jelly was used to anesthetize the eye. Using sterile precautions, the eye was scrubbed with a povidone-iodine 5% solution. The eyelashes were covered with the povidone-iodine solution and a drop of this solution was also placed in the fornix. A sterile lid speculum was then used to hold the eyelids open to expose the conjunctiva. A 1.0-mL syringe attached to a 27-gauge needle was filled with 1.0 cc of 40 mg/mL triamcinolone acetonide. The drug was then injected into the superotemporal sub-Tenon's space forming a bleb of medication; when the anatomy didn't allow for this, it was injected inferiorly.

All patients that underwent STTR were experiencing medically uncontrolled ocular hypertension. The criteria for clinically significant SIOH (CSSIOH) were one or more of the following: (1) an IOP >25 mm Hg; (2) ≥10 mm Hg increase of IOP from baseline; (3) ≥5 mm Hg in crease of IOP from baseline to raise the IOP >21 mm Hg.

The procedure of STTR was as follows (Figure 1): The patient received topical lidocaine to numb the subject eye and was placed in the supine position. The eye was prepped and draped in the usual sterile fashion for vitreoretinal surgery. A Lieberman lid speculum was placed. Westcott scissors and 0.12 forceps were used to remove the sub-Tenon's steroid depot. The conjunctiva was then reapproximated using 7-0 Vicryl sutures. Subconjunctival injections of cefazolin and gentamicin were given. Topical erythromycin was applied to the eye, which was patched after removal of the lid speculum and drapes.

Figure 1. Intraoperative photos of STTR: (A) Sub-Tenon's triamcinolone acetonide deposit (red arrow) located inferiorly. (B) Sub-Tenon's triamcinolone deposit exposed after conjunctiva opened with 0.12 forceps and Westcott scissors. (C) Depot removed. Note gap in conjunctival tissues denoting removal of STT in its entirety. (D) Conjunctiva approximated together and closed with three Vicryl 7-0 sutures.

Average time to STTR was also calculated. A reassessment of mean IOP, VA, and number of IOPLAs, including mean results, was made at post-STTR follow-up visits consisting of one week (post-STTR one wk), four weeks (post-STTR four weeks), and 12 weeks (post-STTR 12 weeks).


The demographic and diagnostic data for each of the 25 eyes studied are summarized in Table 1. Mean age of the study population was 58 years (range 18 to 92 years). Seven of the subjects were males (32%) and 15 were females (68%). Eighty-eight percent of the subjects were Caucasian and the remaining 12% were African American. Five patients had a history of steroid response. The following were the respective causes of the CME encountered in the study population: Irvine-Gass Syndrome (36%), anterior uveitis (36%), epimacular membrane (20%) and central retinal vein occlusion (8%). At baseline, none of the subjects met the clinical criteria for CSSIOH as depicted in Table 2. Twenty-four percent of the eyes were prescribed IOPLAs. Five eyes were prescribed two IOPLAs and one eye a single IOPLA.

Mean IOP measurements at baseline, post-STT max, and post-STTR 12 weeks were 16.4 (±0.7), 38.5 (±1.4), and 15.1 (±0.8) mm Hg, respectively (Table 3). The range in IOP at post-STT max was found to be 28-50 mm Hg. The average time to develop this maximum increase in IOP after STT was 21 weeks. All eyes had clinically significant ocular hypertension at the time of post-STT max according to one or more criteria, as displayed in Table 2. After STTR, IOP returned approximately to baseline levels by the time of post-STTR four weeks follow-up visit. All 25 STTR surgical cases were without complications. The average time between STT and STTR was 28 weeks.

Along with this increase in IOP after STT, a statistically significant increase in both the number of IOPLAs prescribed (P<.001) and the number of eyes receiving IOPLAs (P<.001) were also observed. The average number of IOPLAs at base-line was 0.44 (Table 4). At the time of post-STT max, the required number of IOPLAs had increased to 3.16. Post-STTR 12 weeks showed a decrease in the average number of IOPLAs (1.16) compared to post-STT max. The percentage of patients on IOPLAs at the following times — baseline, post-STT max, and post-STTR 12 weeks — were 24%, 100%, and 63%, respectively (Table 5).

Average visual acuities at baseline, post-STT best, and post-STTR 12 weeks were 20/139, 20/47, and 20/149, respectively (Table 6). The improvement in VA after STT was statistically significant (P<0.001). Average time to achieve post-STT best was 17 weeks.

Figures 2A-D provide pairwise comparisons of: (a) baseline levels of each variable to the post-STT best/post-STT max and to follow-up examination levels (right-side); and (b) the post-STT best/post-STT max to follow-up examination levels of each variable (left-side).

The following summary information is provided for each pairwise comparison: qualitative significance of any change (wanted result/unwanted result/no change), clinical description of any change (eg, “increase in IOP level”), and the statistical significance of any change (red = significant at P<.05/black = nonsignificant at P>.05).

As depicted in Figure 2A, there was a statistically significant improvement in VA at post-STT best and post-STTR four weeks compared to baseline. When comparing post-STTR follow-up visits to post-STT best, there is a statistically significant decline in VA. A statistically significant increase in IOP was seen at post-STT max and post-STTR one week compared to baseline. A return to near baseline IOP level at post-STTR 12 weeks was achieved, albeit statistically nonsignificant (Figure 2B). As shown in Figure 2B, there was a statistically significant decrease in IOP after steroid depot removal. There is a significant decline in the number of IOPLAs prescribed (for people and eyes) between the post-STT max and follow-up visits; however, the number of IOPLAs prescribed at follow-up does not return to baseline levels. There is a significant decline in the number of subjects (and eyes) prescribed IOPLAs (ie, on/off IOPLA) between the post-STT max and follow-up visits; however, the number of subjects (and eyes) on IOPLAs at follow-up also does not return to baseline levels.


Clinically significant SIOH is a well-established complication of periocular steroid injections. Our series demonstrated the reversibility of CSSIOH after STTR based on decreased IOP and reduced number of IOPLAs over time, indicating STTR has a significant palliative effect in reducing CSSIOH. To our knowledge, this represents the largest series of this nature. Although mean IOP at final follow-up returned to levels similar to those at STT (baseline), a complete return to baseline did not occur as manifested by a greater number of IOPLAs at final follow-up (post-STTR 12 weeks). Given the continued statistically significant increase in the number of IOPLAs for patients throughout the follow-up period, we cannot definitively conclude that STTR can fully reverse the effects of CSSIOH. It is our hypothesis that this may be partially due to conservative decision making on the clinicians' part, preventing full withdrawal of all IOPLAs in order to avoid rebound ocular hypertension. However, this hypothesis can only be confirmed within the context of a formal prospective trial that includes the withdrawal of IOPLAs following the normalization of IOP during follow-up. This case study provides strong evidence for the possibility that STTR may lead to a full recovery of baseline IOP and argues for the necessity of such a trial in the near future. It was also evident that the VA benefits conferred by STT also regressed subsequent to STTR. This was due to recurrence of CME in all patients in whom VA worsened, as demonstrated by fluorescein angiography (Figure 3). The relatively high percentage of subjects in this series with anterior uveitis (36%) largely accounts for this tendency.

Figure 3. Late-phase angiography of patient 15. (A) Image taken before injection of STT depot (baseline). Clear evidence of macular edema is appreciated. (B) Resolution of macular edema after STT (post-STT best). (C) At final follow-up visit (post-STTR 12 weeks) after removal of the STT depot.

The mechanism by which SIOH regresses after STTR is likely to be associated with a simple reversal of the cellular processes that have been shown to contribute to SIOH. Downregulation of the MYOC gene upon the removal of the inciting agent may occur, resulting in decreased myocilin production. It is reasonable to assume that lower levels of myocilin and polymerized glycosaminoglycans in the juxtacanalicular tissue of the TM would facilitate aqueous outflow and thus lower IOP.

The decision to proceed with STT therapy to treat CME is often one that is made with some pause by clinicians, as other viable treatment options do exist. Topical and oral nonsteroidal anti-inflammatory medications [NSAIDs] (which are devoid of glaucomatous side effects), as well as topical and oral steroid medications (which can be discontinued at will as unwanted side effects arise), are frequently used with varying degrees of success.16-18 Intravitreal bevacizumab (Avastin, Genentech) and pars plana vitrectomy have also been shown to be beneficial for treating CME in some series.19-22

A desire to avoid the inherent risks of intravitreal injection and vitrectomy and/or failure to respond to NSAIDs or other routes of steroid administration are the primary driving forces leading to depot steroid injection. The knowledge that CSSIOH may be reversible via a procedure that is relatively simple may alleviate some of the hesitation with this step in the treatment algorithm. The adjuvant use of IOPLAs may or may not be used to achieve the desired result.

In this era of intravitreal therapy, when glucocorticoid steroids still represent an important option within the clinical armamentarium, it is important to understand the relative risks and benefits of this modality. This series suggests that one could consider using STT for its anti-inflammatory benefits in treating CME, despite the potential for SIOH, since removal of the depot appears to reverse the unwanted IP elevation. Although the VA benefit was not maintained in this series after removal of the steroid depot, even a transient anti-inflammatory benefit is often sufficient to effectively treat some patients. In the appropriate subject, this treatment may be a viable option to control inflammation, CME, and, consequently, VA. However, this must be performed in conjunction with careful IOP monitoring and adequate management of secondary steroid-induced IOP elevations. RP

Miguel A. Busquets, MD, FACS, Nikolai S. Ždrale, MD, and John P. Nairn, MD, practice with Associates in Ophthalmology of the Associates Medical Center in Pittsburgh, PA. Kelley Kidwell and Ji-In Choi are graduate students in the Department of Biostatistics at the University of Pittsburgh Graduate School of Public Health. Richard D. Day, PhD, is associate professor of biostatistics at the University of Pittsburgh. None of the authors reports any financial interest in any products mentioned in this article. Dr. Busquets can be reached via e-mail at


  1. Kuo HK, Lai IC, Fang PC, Teng MC. Ocular complications after a sub-tenon injection of triamcinolone acetonide for uveitis. Chang Gung Med J. 2005;28:85-89.
  2. Herschler J. Intractable intraocular hypertension induced by repository triamcinolone acetonide. Am J Ophthalmol. 1972;74:501-504.
  3. Clark AF, Morrison JC. Steroid-induced glaucoma. Glaucoma: Science and Practice New York, NY; Thieme; 2003:197-206.
  4. Spaeth GL, Rodrigues MM, Weinreb S. Steroid-induced glaucoma: A. Persistent elevation of intraocular pressure B. Histopathological aspects. Transactions of the American Ophthalmological Society. 1977;75:353-381.
  5. Skuta GL, Morgan RK. Corticosteroid-induced glaucoma. The Glaucomas. 2nd ed. II. St. Louis, MO; Mosby; 1996:1177-1188.
  6. Kocabora MS, Yilmazli C, Taskapili M, et al. Development of ocular hypertension and persistent glaucoma after intravitreal injection of triamcinolone. Clin Ophthalmol. 2008;2:167-171.
  7. Kalina PH, Erie JC, Rosenbaum L. Biochemical quantification of triamcinolone in subconjunctival depots. Archives of Ophthalmology. 1995;113:867-869.
  8. Yamamoto Y, Komatsu T, Koura Y, et al. Intraocular pressure elevation after intravitreal or posterior sub-Tenon triamcinolone acetonide injection. Can J Ophthalmol. 2008;43:42-47.
  9. Inatani M, Iwao K, Kawaji T, et al. Intraocular pressure elevation after injection of triamcinolone acetonide: a multicenter retrospective case-control study. Am J Ophthalmol. 2008;145:676-681.
  10. Sapir-Pichhadze R, Blumenthal EZ. Steroid induced glaucoma. Harefuah. 2003;142:137-140, 15
  11. Helm CJ, Holland GN. The effects of posterior subtenon injection of triamcinolone acetonide in patients with intermediate uveitis. Am J Ophthalmol. 1995;120:55-64.
  12. Jonas JB. Intraocular availability of triamcinolone acetonide after intravitreal injection. Am J Ophthalmol. 2004;137:560-562.
  13. Wordinger RJ, Clark AF. Effects of glucocorticoids on the trabecular meshwork: towards a better understanding of glaucoma. Prog Retin Eye Res. 1999;18:629-667.
  14. Johnson DH. Myocilin and glaucoma: a TIGR by the tail? Arch Ophthalmol. 2000;118:974-978.
  15. François F, Victoria-Troncoso V: [Mucopolysaccharides and pathogenesis of cortisone glaucoma (author's transl)]. Klin Monatsbl Augenheilkd. 1974;165:5-10.
  16. Warren KA, Fox JE. Topical nepafenac as an alternate treatment for cystoid macular edema in steroid responsive patients. Retina. 2008;28:1427-1434.
  17. Wittpenn JR, Silverstein S, Heier J, et al. A randomized, masked comparison of topical ketorolac 0.4% plus steroid vs steroid alone in low-risk cataract surgery patients. Am J Ophthalmol. 2008;146:554-560. Epub 2008 Jul 2.
  18. Hariprasad SM, Callanan D, Gainey S, et al. Cystoid and diabetic macular edema treated with nepafenac 0.1%. J Ocul Pharmacol Ther. 2007;23:585-590.
  19. Mirshahi A, Namavari A, Djalilian A, et al. Intravitreal bevacizumab (Avastin) for the treatment of cystoid macular edema in Behçet disease. Ocul Immunol Inflamm. 2009;17:59-64.
  20. Soheilian M, Rabbanikhah Z, Ramezani A, et al. Intravitreal bevacizumab versus triamcinolone acetonide for refractory uveitic cystoid macular edema: a randomized pilot study. J Ocul Pharmacol Ther. 2010 Mar 22. [Epub ahead of print]
  21. Rotsos TG, Moschos MM. Cystoid macular edema. Clin Ophthalmol. 2008;2:919-930.
  22. Kiryu J, Kita M, Tanabe T et al. Pars plana vitrectomy for cystoid macular edema secondary to sarcoid uveitis. Jpn J Ophthalmol. 2003;47:479-483.
  23. Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 1988;75:800-802.