Development and Implementation of a Vitreous Hemorrhage Grading Scale

Development and Implementation of a Vitreous Hemorrhage Grading Scale

Figure 1. The Vitreous Hemorrhage Density Grading Scale: Grade 1

Vitreous hemorrhage (VH) is a significant cause of vision loss worldwide. The incidence of VH has been estimated at 7/100 000 annually.1 VH occurs when normal retinal vessels are torn (posterior vitreous detachment, retinal tears, or trauma), when retinal blood vessels leak or rupture (proliferative diabetic retinopathy [PDR], retinal vascular events, retinal macroaneurysms, or hemoglobinopathies), or in association with bleeding from other vascular systems (age-related macular degeneration [AMD], subarachnoid hemorrhage, or intraocular tumor).2

Posterior vitreous detachment with retinal tear is the most common underling etiology of VH, followed by PDR.1 Visual symptoms of VH range from mild floaters to severe loss of vision, with the diagnosis being made by using either indirect ophthalmoscopy or B-scan ultrasonography.

There is no single accepted standardized scale for grading VH. Therefore, it is largely subjective and tends to vary among clinicians. Some prefer qualitative descriptions (mild/moderate/severe or small/medium/large) while others choose a quasi-quantitative approach (1+ to 4+).3 Significant intra- and inter-observer variability may exist in the use of these grading schemes. A grading system that references reproducible photographs or drawings, analogous to the retinal photos in the Early Treatment of Diabetic Retinopathy Study (ETDRS), will facilitate communication between healthcare professionals.4,5 This may also lead to standardization of care for these patients by allowing for differentiation into the appropriate group.

We propose the following grading scale for VH, based upon visibility of the retina. A specified number of retinal clock hours must be visualized; then the patient is placed in the appropriate category.


Figure 2. The Vitreous Hemorrhage Density Grading Scale: Grade 2

Grade 0: No blood present in the vitreous, the entire retina is visible.

Grade 1: Some hemorrhage present, which obscures between a total of 1 to 5 clock hours of retina. Laser photocoagulation (PRP) can be successfully performed (Figure 1).

Grade 2: Hemorrhage obscures between a total of 5 to 10 clock hours of central and/or peripheral retina, or a large hemorrhage is located posterior to the equator, with varying clock hours of anterior retina visible. Laser is feasible, but a full panretinal photocoagulation (PRP) can not be placed (Figure 2).

Grade 3: A red reflex is present, with no retinal detail seen posterior to the equator, precluding any photocoagulation (Figure 3).

Grade 4: Dense VH with no red reflex present (Figure 4).

One can see the gradient along which the scale is constructed. There are specific grades or categories in which to place the patient, with an obviously small spectrum within each grade, correlating to the amount of hemorrhage in each grade. For instance, Grade 2 has between 5 and 10 clock hours of hemorrhage. Although subdividing each grade into more specific subcategories might make it more precise, this would complicate its ease of use. Note that patients with a retinal tear causing VH may have that tear obscured by hemorrhage, and need to be treated appropriately.


This proposed classification is based not upon the actual amount of blood in the vitreous, but on visibility of the underlying retina. These objective endpoints have been utilized in two major phase 3 clinical trials with excellent clinical and photographic reproducibility.6-8 The clinical trials evaluated and followed patients with VH of over
1 month duration, following an intravitreal injection of ovine hyaluronidase (Vitrase, ISTA Pharmaceuticals, Inc., Irvine, Calif).9 Decrease (or change) of VH, using the proposed VH grading scale, was used as a secondary endpoint in these studies, with 1125 patients evaluated by 137 separate investigative sites over multiple visits. As one of the secondary endpoints in a phase 3 clinical trial, the VH grading scale was validated and found to be reproducible between multiple centers prior to its use in phase 3 trials.

The VH grading scale was used as a secondary endpoint in these safety and efficacy trials. This subjective endpoint had comparable P values to the increase in best-corrected visual acuity (BCVA), an endpoint that was evaluated objectively. Comparison of the saline (control) group and patients receiving 55 IU of intravitreal Vitrase had a P value of <0.001 at months 1 and 2 and <0.004 at month 3, with respect to BCVA. The subjective, or investigator graded reduction assessment in VH, using the proposed scale described in this article, showed a P value of <0.001 for the saline and 55 IU groups at months 1 and 2, with a P value of <0.003 at month 3. The scale was not only used with ease and efficacy by all examiners, but showed reproducibility and correlation with an objective endpoint.

Figure 3. The Vitreous Hemorrhage Density Grading Scale: Grade 3

A standardized density grading scale should also facilitate several aspects of clinical management of VH. As determined by the DVRS, patients with an initial VH are managed via watchful waiting (observation) for up to 3 months.4,5 They may, in fact, be followed by multiple healthcare providers during this time. Certainly a
standardized grading system of VH would be helpful in that initial phase of their treatment. By placing the patient in the appropriate category, it should aid in assessing progress and prognosis, which in turn, may have therapeutic implications. A patient with a VH which has not cleared sufficiently in accordance with the VH grading scale may be deemed a candidate for vitrectomy earlier than had watchful waiting been employed as per the DVRS.4,5 This may result in earlier useful vision. This applies primarily to retinal vasculopathies amenable to thermal photocoagulation or pharmacological intervention, such as diabetic retinopathy or venous occlusive
disease. Retinal tears causing a VH need to be treated appropriately.

A scale based upon retinal visibility will standardize communication among clinicians, with regard to the functional extent of bleeding. In addition, for ease of communication, the clock hours of VH for each grade could be averaged to attain an "average clock hours of vitreous hemorrhage" composite number.

While the definition of a semi-quantitative score of a "2+ vitreous hemorrhage" may vary between clinicians, if a specified amount of retinal detail is required to be visualized in order for a patient to be placed in the appropriate grade, the classification will become standardized. This is especially important as telemedicine, in addition to retinal photography, becomes more prevalent in the diagnosis and treatment of patients.10-14 The first author has used this scale as a standard in communication with other retinal specialists regarding patients traveling between institutions. In addition, the scale has proved useful in resident teaching, as a standard can be employed.


Figure 4. The Vitreous Hemorrhage Density Grading Scale: Grade 4

The current therapeutic options for VH in PDR include observation and vitrectomy. According to the DVRS, a patient with a non- clearing VH for 3 months is a candidate for pars plana vitrectomy (PPV), although some surgeons may operate sooner.4,5,15,16 Although a PPV is usually a technically easy procedure to perform, it is not
without risk.

Most Grade 3-4 hemorrhages are visually significant enough to warrant intervention but not to justify early vitrectomy; therefore, these eyes may be amenable to less invasive emerging pharmacotherapy. Based upon recently reported phase 3 clinical trials6,7 and other publications17,18, Grade 3-4 hemorrhages may benefit from intravitreal injection of ovine hyaluronidase9 rather than extended observation. Vitrase has been shown to statistically reduce VH densities within 1 month of injection in the phase 3 clinical trials.6,7 Advanced grades of VH may have longer clinical courses, poorer visual outcomes, and may benefit from early surgical intervention.16 The Diabetic Retinopathy Vitrectomy Study and others demonstrated that early vitrectomy significantly increased the likelihood of a good visual outcome.4,5

In summary, we present a clinically useful diagnostic scale to standardize the grading of VH based upon specified amounts of retinal detail visible. Secondarily, we discuss its impact on the treatment of the underlying disease process. This scale will be useful in facilitating communication between healthcare professionals, and ultimately, in developing a VH treatment algorithm.


1. Lindgren G, Sjodell L, Lindblom B. A prospective study of dense spontaneous vitreous hemorrhage. Am J Ophthalmol. 1995;119(4):458-65.

2. Spraul CW, Grossniklaus HE. Vitreous Hemorrhage. Surv Ophthalmol. 1997;42(1):3-39.

3. Ziemianski MC, McMeel JW, Franks EP. Natural history of vitreous hemorrhage in diabetic retinopathy. Ophthalmology. 1980;87(4):306-12.

4. The DVRS Research Group. Two-year course of visual acuity in severe proliferative diabetic retinopathy with conventional management. Diabetic Retinopathy Vitrectomy Study (DRVS) Report #1. Ophthalmology. 1985;92(4):492-502.

5. The Diabetic Retinopathy Vitrectomy Study Research Group. Early vitrectomy for severe vitreous hemorrhage in diabetic retinopathy. Four-year results of a randomized trial: Diabetic Retinopathy Vitrectomy Study Report 5. Arch Ophthalmol. 1990;108(7):958-64.

6. Kuppermann BD, Thomas EL, DeSmet MD, Grillone, LR. Pooled efficacy results from two multinational randomized controlled clinical trials of a single intravitreous injection of highly purified ovine hyaluronidase (Vitrase) for the management of vitreous hemorrhage. Am J Ophthalmol. 2005;140(4):573-584.

7. Kuppermann BD, Thomas EL, DeSmet MD, Grillone, LR. Safety results of two phase III trials of an intravitreous injection of highly purified ovine hyaluronidase (Vitrase) for the management of vitreous hemorrhage. Am J Ophthalmol. 2005;140(4):585-598.

8. Bhavsar AR. Vitrase for the treatment of vitreous hemorrhage. Retinal Physician. 2005;2(5):42-44.

9. Vitrase Package Insert. ISTA Pharmaceuticals, Inc. Irvine, CA 2005.

10. Liesenfeld B, Kohner E, Piehlmeier W, et al. A telemedical approach to the screening of diabetic retinopathy: digital fundus photography. Diabetes Care. 2000;23(3):345-8.

11. Stefansson E, Bek T, Porta M, et al. Screening and prevention of diabetic blindness. Acta Ophthalmol Scand. 2000;78(4):374-85.

12. Luzio S, Hatcher S, Zahlmann G, et al. Feasibility of using the TOSCA telescreening procedures for diabetic retinopathy. Diabet Med . 2004;21(10):1121-8.

13. Schneider S, Aldington SJ, Kohner EM, et al. Quality assurance for diabetic retinopathy telescreening. Diabet Med. 2005;22(6):794-802.

14. Lipp E. Telemedicine: will it ease the burden on the retina community? Retinal Physician. 2006;3(2):47-52.

15. Thompson JT, de Bustros S, Michels RG, Rice TA. Results and prognostic factors in vitrectomy for diabetic traction-rhegmatogenous retinal detachment. Arch Ophthalmol. 1987;105(4):503-7.

16. Cordido M, Fernandez-Vigo J, Fandino J, Sanchez-Salorio M. Natural evolution of massive vitreous hemorrhage in diabetic retinopathy. Retina. 1988;8(2):96-101.

17. Bhavsar AR, Small KW. Reducing the impact of vitreous hemorrhage. Retinal Physician. 2006;3(2):S1-S8.

18. Lieberman, RM, Zatezalo. Use of Vitrase for vitreous hemorrhage in non-surgical candidates. ARVO 2006. Accepted for presentation.

Ronni M. Lieberman, MD, is Assistant Clinical Professor of Ophthalmology at the Mount Sinai Medical Center, Department of Ophthalmology, New York, NY. Dr. Lieberman is a consultant for ISTA Pharmaceuticals, Inc.

J. A. Gow, MD is affiliated with Medical Affairs at ISTA Pharmaceuticals, Inc., Irvine, Calif.

Lisa R. Grillone, PhD, is Vice President, Clinical Research and Medical Affairs at ISTA Pharmaceuticals, Inc.