This study aimed to evaluate risk factors and incidence for postoperative IOP elevation after vitreoretinal surgery with silicone oil tamponade in eyes without a previous history of glaucoma or ocular hypertension, and a predictive model was developed based on this, in the hope of early detection and treatment of the patients concerned. Preexisting glaucoma is considered a very important risk factor for IOP elevation after PPV and SO tamponade. It had described that approximately 5.9%-7% of patients with elevated IOP after surgery had a history of preoperative glaucoma or ocular hypertension in previous studies.15,16We excluded patients with the history of glaucoma or ocular hypertension from our study to detect risk factors other than it. Patients most often have elevated IOP in the first week after surgery. During this period, patients should be closely monitored and followed up, and appropriate treatment should be taken to control IOP at an early stage to achieve better treatment and prognosis.
In previous studies, risk factors for IOP elevation after PPV combined with silicone oil tamponade include diabetes, aphakic eye, and silicone oil viscosity.17,18 In our study, single factor analysis revealed that age, myopia, retinal detachment, macular hole, biological parameter significant were risk factors for elevated IOP after PPV. Moreover, in the prediction model, we further found that hypertension, diabetes, type of IOL, the use of intraoperative laser and silicone oil viscosity can also influence the development of postoperative IOP elevation.
Our results indicated that younger patients have more tendency to develop high IOP postoperatively. This finding is comparable with Pillai et al19 which considered that the increased incidence of trabeculitis and anterior chamber inflammation may account for the high IOP symptoms in younger patients after surgery because they are more likely to develop an inflammatory response. However, there is no clear evidence to confirm this conclusion in our study, which requires further studies to be conducted subsequently.
Although biological parameters significantly affected the postoperative IOP in our study, few studies evaluated the effect of biological parameter such as AL, ACD, CCT, LT and WTW for IOP elevation after vitreoretinal surgery with silicone oil filling. Previous studies had found a positive correlation between CCT and IOP values.20–22The tendency of higher IOP readings with longer axial length had also been reported .23 Meanwhile, Hoffmann et al.24 highlighted the high correlation of IOP measures with thicker central cornea, a thicker lens, and a longer posterior segment length. WTW was positively correlated with ACD,25which can explain that WTW has a positive association with IOP. These studies only reported the correlation between biological parameters and intraocular pressure, did not analyze the impact on the occurrence of elevated IOP after vitrectomy and silicone oil tamponade. Our study is the first to discuss this factor and finds a high positive correlation between postoperative IOP elevation and AL, ACD and WTW, as well as a negative correlation with LT. The correlation between biological parameters and IOP needs to be further explored and explained subsequently
Many articles have reported that aphakic is a strong risk factor for IOP elevation after PPV and SO tamponade,15,26this result is demonstrated in our article as well. We also found that patients with pseudophakia is also a risk factor for IOP elevation. Chang27 first presented that the main reason for IOP elevation in vitrectomized eye was the diffusion of oxygen from the vitreous cavity to the anterior chamber. This caused alterations in the trabecular meshwork leading to reduced aqueous outflow with increase in IOP and glaucoma.28,29Since the crystalline lens contains proteins which metabolize oxygen,30,31it has the potential to reduce the oxidative stress on the trabecular meshwork and prevent the oxidative damage. In addition, it also acts as a barrier to prevent oxygen from entering the anterior chamber. This could explain why the model estimates that lens status is one of the major factors influencing postoperative IOP elevation.
Comparing with PPV alone, previous studies32–34 agreed that combined phacoemulsification and intraocular lens implantation (PE & IOL) seems to exert an additive effect on transient IOP elevation after PPV, which shows the same results with our prediction model. Then the long-term IOP lowering effect of cataract surgery may lead to decreased IOP in the late postoperative period. This may be due to cataract surgery inducing more inflammation in the trabecular meshwork, leading to elevated IOP in the first few days after surgery, while PPV combined with cataract surgery may also cause anterior segment inflammation and disruption of the blood-water barrier.26,35−37The long-term IOP reduction after cataract surgery has been explained by the improvement of outflow facility, deepening of the anterior chamber and posteriorization of the lens-iris diaphragm after cataract surgery.38,39
Patients with RD or MH were strongly associated to develop an IOP rise in our study. Fujikawa et al. 40reported the same conclusion that MH group was significantly at high risk of IOP increase after vitrectomy. The result of recent study41 had suggested that because the vitreous lacks a diffusion barrier for the oxygen, the partial oxygen pressure is highly elevated in the vitreous cavity. For patients with RD and MH, we performed a more thorough vitrectomy removing the peripheral vitreous completely but core vitrectomy for other patients, it would increase oxygen stress to the trabecular meshwork42 in eyes with RD which may cause the IOP elevation.
Although features such as hypertension, diabetes, types of IOL, intraoperative laser and silicone oil viscosity were not statistically significant in the univariate analysis, their importance in the prediction model accounted for a certain percentage. Due to the fact that postoperative IOP elevation is the result of a multifactorial interaction and a common univariate analysis is rather homogeneous, whereas the prediction model performs the analysis as a whole therefore yielding a more comprehensive risk factor.
Silicone oil viscosity may interact with the lens, lower silicone oil viscosity is more likely to result in silicone oil emulsification into the anterior chamber in patients with aphakic or IOL eyes, leading to high postoperative IOP. In patients undergoing vitrectomy combined with cataract ultrasound emulsification and IOL implantation, IOL is risk factor due to its weak barrier effect and inability to consume oxygen, therefore in the prediction model, different types of lens also influence the development of postoperative IOP elevation. The roles of diabetes and hypertension in IOP elevation had also been discussed in many studies, but the conclusion remains controversial, further observation and analysis are necessary in the follow-up study.
This study has several limitations. The retrospective nature of the study means it contains some bias due to poor patient adherence and incomplete data. We missed the specific IOP values of some patients, basing on the doctor prescribed intraocular pressure lowering drugs to determine whether the patients have postoperative high IOP, which will also lead to statistical errors. Meanwhile, although the prediction model is interpretable, the accuracy is still weak, an updated algorithm is needed to build a more reliable model.
IOP elevation is one of the most common and severe complications of PPV with silicone oil tamponade in eyes. If a patient's high IOP is not effectively controlled after surgery, the clinical option is usually to control IOP by using IOP-lowering eye drops or removing intraocular silicone oil early to prevent optic nerve damage, but the early removal of silicone oil may lead to secondary problems such as recurrent retinal detachment. Therefore, it is clinically relevant to investigate the risk factors that may lead to postoperative hypertension and to take interventions to reduce the incidence of postoperative hypertension accordingly. Also, due to the high prevalence of myopia and cataract in China, our study innovatively included biological parameters as factors for a systematic analysis. The larger amount of data in this study allows for a more comprehensive and accurate analysis of the risk factors that may be associated with postoperative high IOP. Meanwhile, the statistical results and prediction model can provide a basis for our subsequent prospective study.