Introduction

Raised intraocular pressure (IOP) is a common postoperative complication of vitrectomy. Topical steroids and antibiotics are often administered after a vitrectomy to reduce inflammation and risk of infection, thereby promoting better healing and visual outcomes. However, the administration of steroids has also been linked with the development of raised IOP [1]. At our local center, we also observed that some patients developed high IOP accompanied by severe eye pain, headache, nausea, and vomiting following the administration of topical steroids. Elevated IOP usually normalizes within a few days after discontinuing the steroid treatment. However, raised IOP can result in optic nerve damage and potentially irreversible vision loss [2]. Factors such as a family history of open-angle glaucoma, myopia, connective tissue disorders, and age have been linked with the development of raised IOP [3].

In recent years at our center, we have replaced the 20-gauge (20G) vitrectomy system with the 23-gauge (23G) system. Compared to the 20G system, the smaller size of the 23G surgical instruments results in less invasive surgery, faster healing, reduced postoperative discomfort, and fewer complications. As a result, the 23G could potentially reduce the need for administering steroid treatment to reduce inflammation post-surgery [4]. Although several studies evaluated the incidence of high IOP after vitrectomy, few studies assessed the risk factors that lead to steroid-induced ocular hypertension (SIOH). Therefore, in this study, we aimed to evaluate the incidence of SIOH following 23G vitrectomy. In addition, we also aimed to identify the risk factors for developing SIOH after 23G vitrectomy and the impact of early steroid discontinuation on IOP in patients at risk of developing SIOH.

Methods

Results

Risk factors for SIOH

The univariate analysis identified myopia, preoperative vitreous hemorrhage, cryotherapy, silicone oil, and gas filling as risk factors for developing SIOH (IOP ≥ 23 mmHg) and myopia, cryotherapy, and silicone oil filling as risk factors for developing severe SIOH (IOP ≥ 30 mmHg) (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Univariate logistic regression of risk factors for SIOH in vitrectomy patients.

https://doi.org/10.1371/journal.pone.0347527.t001

The multivariate analysis identified myopia (odds ratio (OR) 5.22) and silicone oil filling (OR 8.20) as independent risk factors for SIOH. For severe SIOH (IOP ≥ 30 mmHg) myopia and silicone oil filling were also identified as risk factors with an OR of 3.23 and 12.86, respectively (Table 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Multiple logistic regression of risk factors for SIOH in vitrectomy patients.

https://doi.org/10.1371/journal.pone.0347527.t002

Comparison between the traditional and adjusted steroid treatment

There was no significant difference in age, preoperative IOP, the proportion of myopia, and the number of cases with silicone oil filling between the two treatment groups at baseline. The patients in the traditional steroid group received an average of 9.80 ± 3.31 days of topical steroids and 0.37 ± 0.89 days of systemic steroids. The patients treated with adjusted steroid therapy received on average a shorter treatment (7.09 ± 3.14 days) of topical steroids and a longer treatment of systemic steroids (1.36 ± 1.77 days) (Table 3). After excluding patients who discontinued topical steroids due to SIOH, the myopic patients in the traditional steroid group received significantly more days of topical steroid treatment (10.86 ± 2.59 days versus 6.42 ± 2.96 days (P = 0.004)). Similarly, the patients with silicone oil tamponade in the traditional treatment group received more days of topical steroid treatment when compared to the adjusted steroid group (10.88 ± 2.55 days and 7.03 ± 3.06 days P < 0.001).

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Characteristics of the study population before and after control medication duration.

https://doi.org/10.1371/journal.pone.0347527.t003

As shown in Table 4, when compared with the traditional steroid group, the incidence of SIOH was significantly lower in the adjusted steroid group for both IOP of 23 mmHg or higher (26.37% versus 17.11%, P = 0.010 and IOP of 30 mmHg or higher (16.92% versus 9.14%, P = 0.007). The adjusted steroid group also had a longer mean average onset for SIOH than the traditional steroid group (5.26 ± 1.90 versus 4.43 ± 1.80 days P = 0.02).

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. SIOH related factor before and after changing the topical steroid administration.

https://doi.org/10.1371/journal.pone.0347527.t004

Discussion

SIOH is a common side effect following vitrectomy. Although the IOP of most patients with SIOH can be restored to normal levels by stopping steroid therapy, studies have shown that even a short-term state of high IOP may cause irreversible damage to the optic nerve and other tissues [6,7]. The introduction of new more accurate surgical techniques may reduce the need for post-operative steroid treatment. Therefore, in the study, we aimed to identify the risk factors for developing SIOH following vitrectomy to optimize the post-operative steroid dose for these patients.

In our study, 20.56% of patients experienced SIOH and nearly 27.9% of the patients with SIOH experienced an IOP of 40 mmHg. Some of the SIOH patients suffered from significant eye pain accompanied by headache, vomiting, and other symptoms. Moreover, some patients developed asymptomatic elevated IOP, which was detected during routine follow-up visits. Although their IOP returned to normal later, glaucomatous optic disc atrophy was found months later in the clinic work. Cheng et al. [8] reported an SIOH incidence rate of 14.6% following a 20G vitrectomy, while Brennan et al. reported a 22% elevated IOP incidence rate within two weeks after a 23G vitrectomy [9]. The slight difference in the SIOH incidence rate compared to that of our study might be due to different surgical techniques and definitions used to define high IOP. However, consistent with previous work, our findings highlight the importance of monitoring the IOP of patients treated with steroids to minimize the risk of damaging the eye [6,7].

The onset and duration of raised IOP varied between studies. Roberti et al. noted that the onset of raised IOP can occur within a few hours to several years after chronic steroid use [2]. The IOP usually returned to normal within 1–4 weeks after stopping the medication [2]. Cheng et al. [8] reported that SIOH occurred on average within 12 days post-surgery (range 4–20 days). In this study, cases of SIOH occurred 2–10 days after surgery, with about 83.78% of the cases occurring between the third and seventh day after surgery. In our study, we did not observe an onset of SIOH after 10 days post-surgery since the steroid treatment was discontinued 2 weeks after surgery. However, compared to previous studies, we observed a more rapid increase in the IOP within 24 hours. The differential response to steroids in our study may be attributable to genetic variations that influence the patients’ sensitivity to steroids [10], or potentially to other, as yet unidentified, surgery-related factors.

Consistent with previous work, in our study, myopia was identified as a potential independent risk factor for SIOH [2,3,10]. Steroids may inhibit the trabecular meshwork cells’ ability to phagocytose cellular debris, leading to debris blockage and subsequent obstruction of aqueous humor outflow and ultimately raised IOP [2,10]. Cho et al. found that individuals with lower trabecular meshwork height (especially <646.75 μm) are more prone to SIOH [11]. Myopic patients tend to have a lower trabecular meshwork height and a larger Schlemm's canal [12]. Therefore these patients are more at risk of developing IOP.

Silicone oil tamponade is one of the major causes of elevated IOP after vitrectomy [13]. Although the mechanism behind the development of raised IOP in patients treated with silicone oil remains unclear, several confounding factors can increase the risk of developing SIOH. The silicone oil tamponade can cause inflammation and pupillary block. In addition, the silicone oil may leak into the anterior chamber and increase the IOP. In the late postoperative stages, silicone oil emulsification [14], anterior chamber angle closure [15], and intraocular inflammation may further increase the risk of developing SIOH [16]. Previous studies have confirmed that both standard and heavy silicone oil can lead to intraocular inflammation [17]. Histopathological analysis of retinal samples indicates that silicone oil may cause a delayed type 4 hypersensitivity reaction [17]. Patients treated with silicone oil tamponade are more likely to have severe retinal detachment that requires a longer surgical duration and recovery time. Moreover, following the surgical procedure, the patient has to lie in the prone position to ensure the proper positioning of the silicone oil, enhance healing, and reduce the risk of complications. However, the prone position may increase the risk of developing IOP [18]. In the adjusted steroid regimen, despite increased systemic steroid use, the shortened duration of topical steroids was associated with a significantly lower incidence of SIOH. It is well established that both systemic and topical corticosteroids can elevate intraocular pressure [19]. However, the risk of IOP elevation is generally lower with systemic administration than with topical ocular use [20]. Moreover, systemic steroid-induced ocular hypertension tends to occur after prolonged use, with onset typically ranging from weeks to months [21], whereas topical steroid-induced IOP elevation often develops within days to a few weeks. Given the relatively short observation period in our study, the contribution of systemic steroids to IOP elevation may have been limited. Therefore, the observed reduction in SIOH incidence in the adjusted group likely reflects the dominant effect of shortened topical steroid exposure, outweighing any potential risk from concurrent systemic steroids. Nevertheless, due to the retrospective design and the concurrent use of both routes, the independent effect of systemic steroids on IOP could not be fully quantified. The inter-individual variability in steroid response observed in our study may be influenced by genetic predisposition, ocular structural characteristics, or other population-specific factors. For example, the high prevalence of myopia in East Asian populations, a key risk factor identified in our study, may partly account for the observed risk profile. Additionally, genetic variants associated with steroid sensitivity differ across populations; the N363S polymorphism, which is linked to heightened glucocorticoid sensitivity in Europeans, is extremely rare in East Asian population [22]. Further studies incorporating East Asian-specific genetic and anatomical analyses are warranted to elucidate the mechanisms underlying steroid-induced ocular hypertension in this population.

This study has some limitations that have to be acknowledged. The diagnosis of myopia was not determined by the axial length of the eye, which may have resulted in the omission of a small number of patients with mild to moderate myopia, making the estimated odds ratios for myopia conservative. The 23G vitrectomy, as a suture-free incision technique, may have a higher risk of early scleral incision leakage compared to the 20G suture-tightened incision technique [23], which may affect the accuracy of early postoperative IOP measurements. However, some studies have found no statistical difference in IOP between sutured and non-sutured 23G surgeries [9]. In addition, because some follow-up data were collected via telephone and relied on patients’ recall rather than objective measurements, the possibility of recall bias and missed detection of asymptomatic IOP elevation may have been underestimated. Furthermore, given the retrospective nature of this study, causality cannot be definitively established, and the observed IOP elevations in silicone oil-filled eyes may represent a combination of steroid and silicone oil effects. Finally, this study defined postoperative raised IOP as exceeding 23 mmHg, which may have missed some patients with low baseline IOP. It should also be noted that the definition of SIOH varies in the literature; our use of ≥23 mmHg may affect comparability with studies using a higher cutoff (e.g., ≥ 25 mmHg). Our study presents preliminary observations suggesting that shortening the duration of corticosteroid therapy based on postoperative inflammation levels significantly reduces the occurrence of elevated intraocular pressure. Further randomized controlled trials are needed to confirm this clinical finding.

Conclusion

Myopia and silicone oil filling were identified as potential independent risk factors for SIOH after vitrectomy. In addition, reducing the duration of topical steroid therapy was associated with a lower incidence of SIOH, particularly in high-risk patients. Further prospective studies are warranted to validate these findings.

Acknowledgments

We would like to thank the ophthalmology medical staff involved in patient management, including surgeons, nurses, and others. We acknowledge TopEdit LLC for the linguistic editing and proofreading of this manuscript.