Etiology and surgical outcomes in non-resolving vitreous hemorrhage: A prospective study

INTRODUCTION

Vitreous hemorrhage (VH) is defined as bleeding into the space outlined by the internal limiting membrane of the retina posteriorly and posterolaterally, the non-pigmented epithelium of the ciliary body laterally and the zonule and posterior lens capsule anteriorly.^[1] The incidence is approximately seven cases/100000 population.^[2] Various causes of VH in adults are proliferative diabetic retinopathy (PDR) (31–54%), vascular occlusions (4–16%), retinal tear (11–44%), and trauma (12–19%).^[1] Other causes are choroidal neovascular membrane, idiopathic polypoidal choroidal vasculopathy, retinal macroaneurysm, subarachnoid hemorrhage (Terson’s syndrome), and neovascularization elsewhere due to vasculopathies. Rarely, blood dyscrasias, valsalva retinopathy, intraocular tumors, and bleeding from vascularized snowbank in pars planitis can also lead to VH.^[3]

Despite recent improvements in ophthalmologic diagnostic examination techniques, VH often presents a diagnostic challenge. The incidence of only posterior vitreous detachment causing VH is not known. Whether treatment with anticoagulants causes VH, is not known. Whether intravitreal anti-vascular endothelial growth factor (VEGF) is beneficial in VH vitrectomy, is not clear. There is a lack of consensus regarding the timing of surgery in VH. The study was undertaken to find out the etiological factors for non-resolving VH and the outcome of treatment in the form of visual acuity, intraocular pressure (IOP), and any further required interventions.

MATERIAL AND METHODS

This is a prospective interventional study done at a referral eye institute from January 2018 to December 2019. The Institutional Ethics Committee approval was obtained before undertaking the study. The study adhered to the tenets of the Declaration of Helsinki.

Sample size calculation: In January–December 2015, out of 48494 patients, there were 260 cases of VH, of which 145 needed vitrectomy. Hence, in a population of 100, with a confidence level of 95% and a confidence interval of 9%, a sample size of 55 or more was calculated for the study.

Diagnosis and intervention were done by a single trained VR surgeon, thereby limiting the observer and surgeon variation. Sample size calculation of 55 or more was done with a confidence interval of 9% and a confidence level of 95%. The inclusion criteria were patients aged more than 10 years, presenting with dimness of vision for more than 3 months, and who were diagnosed as non-resolving VH. The exclusion criteria were any history of previous ocular surgery and any complications after vitrectomy that needed additional intervention.

At the initial visit, patients were asked about symptoms, previous ocular diseases, and general health. Ophthalmologic examinations included visual acuity with the logarithm of the minimum angle of resolution (logMAR) chart, applanation tonometry, slit-lamp examination with biomicroscopy, and indirect ophthalmoscopy. The density of the hemorrhage was evaluated ophthalmoscopically by estimating the brightness of the red fundus reflex and by grading the visibility of the retina. B-scan ultrasonograms were done. The examination was performed by ophthalmologists trained in ultrasound analysis. Blood tests included hemoglobin concentration, white blood cell count, erythrocyte sedimentation rate, blood glucose level, bleeding time, and clotting time. Serum urea and creatinine were checked. Electrocardiogram (ECG) was performed for all cases and cardiological clearance was sought for whenever needed. Seropositivity was checked for HIV and Hepatitis-B surface antigen (HbsAg). Blood pressure was measured in all cases. After being diagnosed as non-resolving VH and advice for surgery, counseling was done, and a consent form was signed. All patients underwent standard 25G pars plans vitrectomy (7500 cut rate) with posterior hyaloid detachment in the constellation vitrectomy system with Volk Mini Quad XL lens, wide-angle viewing system with stereoscopic diagonal inverter with video recording of each procedure. All patients were informed about the risks and benefits of the surgery, and written informed consent was obtained. Patient was followed up at 1, 5, 14, 25, and 45 postoperative days. A serial colored fundus photograph was taken.

RESULTS

There were 61 eyes of 59 patients from January 2018 to December 2019 of non-resolving VH, who underwent vitrectomy in this study. The data followed normal distribution as shown by a one-sample Kolmogorov–Smirnov test. Male: female ratio in our study was 48:11. The age distribution where the number of VH was maximum was in the group of 41–50 years [Figure 1]. Mean age of presentation for males was 50.60 ± 15.01 years and for females was 54.27 ± 8.55 years. Right eye: left eye ratio was 41:20. Dimness of vision was the main presenting complaints in 81.96% (n = 50 eyes), followed by trauma in 8.19% (n = 5 eyes) and pain and dimness of vision in 3.27% (n = 2 eyes). There was no complaint in 1.6% (n = 1 eye) and floater in 1.6% (n = 1 eye).

Close

Figure 1:

Age distribution of patients with number of eyes.

Export to PPT

The mean duration of presentation after appearance of symptoms was 67.96 ± 79.29 months [Figure 2]. Most cases (n = 24 eyes) presented within 1 month of symptoms. Three patients presented with symptoms having more than 3 years. Post-operative visual acuity remained the same in these cases.

Close

Figure 2:

Duration of presentation from onset of eyes.

Export to PPT

There were 20 eyes (32.7%) for which no systemic cause for VH was found. Diabetes mellitus was present in 45.7% patients (n = 27), which was the major etiology for VH. The number of patients presenting with VH in relation to the type of diabetes mellitus and the duration of diabetes is given in Table 1. Intravitreal anti-VEGF pre- or intraoperatively was not used in any PDR cases undergoing surgery. Systemic causes of VH encountered in this study population are enumerated in Table 2. Ocular causes of VH found in this study population are detailed in Table 3.

Cataract was present in 31.1% eyes (n = 19). One case had undergone phacoemulsification with IOL implantation along with vitrectomy. All cases underwent 25-gauge microincision vitrectomy with endolaser with tamponade as the primary procedure under local or general anesthesia. None of them were asked to stop any anticoagulants if they were taking them before surgery.

Mean pre-operative visual acuity was 1.21 ± 0.44 logMAR, and mean post-operative visual acuity was 0.66 ± 0.54 logMAR, which was highly significant (P = 0.000 paired t-test) [Figure 3].

Close

Figure 3:

Visual acuity at the end of the follow-up time compared with the initial visit. The visual acuity at the 45^th post-operative day is inclined more toward the linear BCVA signifying improved visual acuity. BCVA: Best corrected visual acuity.

Export to PPT

About 26.2% eyes (n = 16) experienced IOP spikes postoperatively, which were treated with anti-glaucoma medications [Figure 4]. About 9.8% eyes (n = 6) had to undergo re-vitrectomy.

Close

Figure 4:

Intraocular pressure variation in the preoperative, 1^st, 5^th, 14^th, 21^st and 45^th postoperative day.

Export to PPT

DISCUSSION

The course of VH is spontaneous clearance slowly at the rate of 1%/day if no recurrent bleed occurs. If the VH does not clear, there may be hemosiderosis bulbi, fibrovascular proliferation, and glaucoma (hemosiderotic, hemolytic, and ghost cell).^[4]

The percentage of cases that receive vitrectomy in VH cases ranges from 0.20 to 0.27 during 2001–2010.^[5] The incidence of VH increased significantly in subjects aged between 40 and 59 years in a recent study and several other studies in the literature.^[5]

In this study, we also found that the mean age of presentation for males was 50.60 ± 15.01 years and for females were 54.27 ± 8.55 years. The incidence of VH was shown to increase significantly with age due to the increasing incidence of the causes of VH in these ages, such as PDR, retinal vein occlusion, and the associated systemic diseases. Right eye was more frequently involved in 67.21%. Dimness of vision was the main presenting complaints in 81.96% (n = 50 eyes), mainly because cases present only when visual disability predominates. Three of our patients presented after 3 years of the appearance of symptoms. These three cases did not show any improvement in visual acuity postoperatively. There are several reports of visual improvement even after 5 years of non-clearing diabetic VH, when vitrectomy was first performed for VH.^[6-8] However, our cases did not show a favorable outcome due to advanced disease pathology.

The causes of VH may depend on the study population and tend to vary with the mean age of the patients as well as the region in which previous studies were performed.^[4,9,10] Proliferative diabetic retinopathy (19.1–54.0%) and retinal break (27–37.3%) were the most common causes of VH in previous studies. Vitreous detachment with or without a tear was only reported as the most common cause of VH in a case series by Lindgren et al.^[4] Majority of the patients were diabetic 45.7% (n = 27) in our study, and PDR accounted for the maximum VH in 42.6% (n = 26 eyes), which is consistent with the other studies.^[4,5]

As most of the cases were in the elderly age group, there was cataract in 19 eyes (31.1%) but only in one case vitrectomy had to be performed combined with a cataract surgery. This could be attributed to better technological advances for vitrectomy along with a better viewing system.

The post-operative visual acuity was significantly better (P = 0.000) in all the cases. Sixteen (26.22%) of patients had improved visual acuity postoperatively to 0.2-00 logMAR.

There were no major post-operative complications except for a few IOP spikes (n = 16 eyes), which were controlled by medication, as was also in a recent study by Han et al.^[11] The re-vitrectomy was for VH in diabetics.

A recent study found that taking aspirin, clopidogrel, or warfarin was more likely to result in VH as a result of an acute posterior vitreous detachment.^[12] This finding suggests that regardless of the type of disease requiring anti-coagulation medication or the disease cause, the patient would have a higher risk for VH. None of the patients in this study were asked to stop any anticoagulants if they were taking before surgery. We did not find higher incidence of VH in patients taking anticoagulants, probably because the design and purpose of the study were different from the earlier studies, and this was the limitation of the study. Another limitation of the study is that, it is a single-center study involving a small population size. A population-based clinical study may add to the findings of this study.

Diabetic Retinopathy Clinical Research Network found no difference between intravitreal ranibizumab and intravitreal saline injection for eyes with VH of severity that precluded PRP and which was the proportion of patients requiring vitrectomy. Therefore, at present, intravitreal anti-VEGF does not seem to have a significant role in the treatment of diabetic VH other than as a pre-operative adjuvant to vitrectomy.^[13] In this study, intravitreal anti-VEGF pre- or intraoperatively was not used in any PDR cases undergoing surgery.

CONCLUSION

Non-resolving VH can be one of the major causes of profound visual loss. It can occur with variable pathologies at any age, gender, and with varied systemic associations. Detailed ocular and systemic history, thorough clinical evaluation, tailor-made laboratory investigation, meticulous and often repeated ocular ultrasonography may guide us for planning vitrectomy and arrive at a probable diagnosis of cause of VH. Micro-incision vitrectomy at the earliest can be done to treat and find out the exact cause of VH. Early vitrectomy can salvage eyes from going irreversibly blind. Careful following of vitreous surgery principles and regard for the underlying pathology for VH gives a favorable outcome.