MINISTRY OF EDUCATION AND TRAININGMINISTRY OF HEALTH HA NOIMEDICALUNIVERSITY NGUYEN DIEU LINH STUDY ON USING INTRAVITREAL BEVACIZUMAB INJECTION AND VITRECTOMY IN THE TREATMENT OF VITREOUS HEMORRHAGE COMPLICATION DUE TO PROLIFERATIVE DIABETIC RETINOPATHY Major : OPHTHALMOLOGY Code : 62720157 MEDICAL DOCTOR DISSERTATION SUMMARY HANOI - 2019 THE DISSERTATION IS COMPLETED AT HANOI MEDICAL UNIVERSITY Scientific guidance: Professor Do Nhu Hon Reviewer 1: Associate professor Tran Thi Nguyet Thanh Reviewer 2: Associate professor Hoang Nang Trong Reviewer 3: Associate professor Nguyen Van Dam The dissertation presented to the Board of Ph.D dissertation at University level at Hanoi Medical University At date month year 2019 The dissertation can be found at: - National Library of Vietnam - Library of Hanoi Medical University LIST OF PUBLIC SCIENTIFIC WORKS RELATED TO THE DISSERTATION Nguyễn Diệu Linh, Đỗ Như Hơn, (2015), “ Evaluation of treatment results for vitreous hemorrhage in diabetic retinopathy after year by intravitreal Avastin injection in combination with vitrectomy”, Vietnamese Journal of Ophthalmology,vol.38, p:37-43 Nguyễn Diệu Linh, Nguyễn Thị Nhất Châu, Đỗ Như Hơn, (2015), “ The effectiveness of intravitreal Avastin injection as an adjunct to vitrectomy in the management of diabetic vitreous hemorrhage”, Journal of Practical Medecine, N0 987, November, p:95-98 INTRODUCTION Diabetic retinopathy is one of the most prevalent and severe ocular disorders, which is a major cause of adult blindness Vitreous hemorrhage ( VH) is a common complication of proliferative diabetic retinopathy ( PDR) Vitrectomy is a predominant treatment for proliferative diabetic retinopathy The incidence of postoperative vitreous hemorrhage after vitrectomy is 20%-60% Bevacizumab (Avastin) is a recombinant humanized monoclonal IgG1 antibody that inhibits human vascular endothelial growth factor (VEGF) Bevacizumab can induce regression of retinal neovascularization in patients with diabetes; therefore, it was suggested that a presurgical administration of intravitreal bevacizumab ( IVB) may reduce intraoperative bleeding during vitrectomy in PDR Recently, numerous studies have reported clinical outcomes of intravitreal bevacizumab as an adjunct to vitrectomy in the management of proliferative diabetic retinopathy In Vietnam, there are no studies that have been performed Therefore, the study named " Study on using intravitreal Bevacizumab injection and vitrectomy in the treatment of vitreous hemorrhage complication due to proliferative diabetic retinopathy” is conducted with the aims of: 1.To evaluate the effect of intravitreal bevacizumab ( IVB) injection before vitrectomy in the management of vitreous hemorrhage complication associated with PDR 2.To analyse some associated factors with the results NEW CONTRIBUTION OF THE THESIS - Intravitreal Bevacizumab injection and vitrectomy in the treatment of vitreous hemorrhage due to proliferative diabetic retinopathy help to limit intraoperative and postoperative complications, improve visual acuity and anatomical results - Systemic factors, complications during surgery not affect treatment results Complications after surgery, additional surgery affect the outcome of treatment STRUCTURE OF THE THESIS The thesis consists of 139 pages: Introduction (2 pages), Overview (36 pages), subjects and methods (24 pages), Results (28 pages), Discussion (46 pages), Conclusion (2 pages), Recommendation (1 page) New contribution (1 page) The thesis consists of 32 tables, charts, pictures, appendix and list of patients CHAPTER OVER VIEW 1.1 DIABETIC RETINOPATHY 1.1.1 Epidemiology of diabetic retinopathy: Diabetic retinopathy ( DR) usually occurs after years of having diabetes After 15 years and 20 years more than 50% and most of diabetic patients have retinopathy, respectively 1.1.2.Pathophysiology of diabetic retinopathy: metabolic changes at the molecular level leading to endothelial cell dysfunction, basement membrane thickening, endothelial cell and pericyte loss cell changes Those abnormalities contribute permeability changes, retinal occlusion, formation retinal hypoxia, which stimulates endothelial, pericyte and retinal pigment epithelium to produce VEGF causing neovascularization 1.1.3.Clinical features of diabetic retinopathy 1.1.3.1 Clinical findings: capillary closure, retinal occlusion, retinal hypoxia, neovascularization New vessels develope on the surface of retina or optic disc, create a widespread neovascular network that can cause complications of vitreous hemorrhage, tractional retinal detachment and neovascular glaucoma 1.1.3.2.Classification of DR: According to Alfédiam’s classification * Non-proliferative diabetic retinopathy: mild, moderate, severe * Proliferative diabetic retinopathy: mild, moderate, severe and complications * Diabetic macular retinopathy: focal, diffuse, cyst, tractional macular edema 1.1.3.3.Complications of PDR *Vitreous hemorrhage: The neovascular tissue grows within the outer cortical vitreous and associate with fibrous tissue proliferation Traction due to posterior vitreous detachment lead to vitreous hemohhrage Classification of vitreous hemohhrage according to MR Romano: Grade 0( no hemohhrage, can see fundus clearly in detail), grade (mild hemorrhage, can see fundus), grade (moderate hemorrhage, can not see the fundus except the disc), grade (severe hemorrhage, unable to see the fundus) *Tractional retinal detachment: Neovascular and fibrous tissue adhering to the posterior hyaloid membrane shrink the retina leading to retinal detachment, possibly with retinal tearing * Neovascular glaucoma: due to response to hypoxia, VEGF diffuses to the anterior chamber, causing neovascular on iris and anterior chamber angle 1.1.4.Therapies 1.1.4.1.Systemic treatment: Treat diabetes, hypertension and kidney disease 1.1.4.2.Panretinal laser photocoagulation: Do laser photocoagulation at peripheral region to decrease VEGF, thereby reducing neovascularization 1.1.4.3.Anti- VEGF *Pegaptanip: connect to VEGF-A 165 branch *Ranibizumab: connect to all kinds of VEGF *Aflibercept: connect to VEGF-A, VEGF-B andplatelet growth factor *Bevacizumab ( Avastin): is a monoclonal immunoglobulin, combining VEGF through antigen-binding sites 1.1.4.4.Vitrectomy *Purpose: Removing vitreous , dissecting and removing fibrous membrane to reattach retina, doing laser photocoagulation to prevent neovascular *Indications: severe vitreous hemorrhage / fibrovascular membrane / traction macular detachment / traction retinal detachment with retinal tear *Technique: vitrectomy from the center to the periphery, making posterior vitreous detachment then segment the fibrovascular membrane by separating them from the retinal surface, re-attach retina, laser photocoagulation, using temponade Combined vitrectomy and phaco surgery to put artificial lenses in cases of cataract * Complication - Complications during surgery: bleeding, retinal tear - Complications after surgery: Hemorrhagic hemorrhage (early, late), retinal detachment, glaucoma and glaucoma, cataract, infection 1.2 BEVACIZUMAB IN OCULAR TREATMENT 1.2.1.Structure: Bevacizumab (Avastin, Genetech Inc., San Francisco, CA) is a monoclonal immunoglobulin, combining VEGF through antigen-binding sites 1.2.2.Pharmacokinetics: The drug exists in vitreous more than 30 days after a dose of 1.25mg / 0.05ml 1.2.3 Mechanism: the drug penetrate the blood-retina barrier, combined with all types of VEGF VEGF inhibitors cause temporary vasoconstriction, new vessels regress completely within 48 hours and maintain for weeks After injection, there is a decrease in both the number and aperture of new vessels that will then develop fibrosis 1.2.4 Indication: to treat proliferative diabetic retinopathy: reduces new vessels of iris, optic disc, retina and reduces vascular leakage, hemorrhage 1.2.5 Side effects 1.2.5.1 Systemic side effects: reduced wound healing, hypertension, myocardial infarction, stroke, even death 1.2.5.2 Ocular side effects after injection (side effects of the drug and injection technique): subjunctival hemorrhage, intraocular inflammation, retinal tear 1.3 THE COMBINATION OF INTRAVITREAL BEVACIZUMAB AND VITRECTOMY TO TREAT PROLIFERATIVE DIABETIC RETINOPATHY 1.3.1.Indications Severe vitreous hemorrhage/ fibrovascular membrane/ tractional macular detachment/ tractional retinal detachment with retinal tear 1.3.2.Researchs on combination therapy: Increased VEGF concentration in vitreous of proliferative diabetic retinopathy eye is a risk factor for failure of vitrectomy Intravitreal injection of anti-VEGF drugs reduces VEGF levels in the vitreous cavity of proliferative diabetic retinopathy eye * Drugs: Ranibizumab and Bevacizumab are rated equally on the effectiveness of treatment and safety The larger molecular size of Bevacizumab should have the advantage of treating proliferative diabetic retinopathy * Dose: the effect of intravitreal injection Bevacizumab for proliferative diabetic retinopathy is not significantly different with the dose ranging from 1.25mg, 2.5mg to 6.2mg Currently the common dosage is 1.25mg / 0.05ml * Time between injection and vitrectomy: In the treatment of proliferative diabetic retinopathy, Bevacizumab is injected before or at the end of the surgery Preoperative Bevacizumab injection is better because the drugs makes it easily to dissect fibrous proliferative membrane, laser photocoagulation, limit bleeding during surgery, postoperative complications * Studies using Bevacizumab (1.25mg / 0.05ml) intraocular injection before vitrectomy: the authors in the world report that intravitreal Bevacizumab injection (1.25mg / 0.05ml) preoperative 1-2 weeks in the treatment of proliferative diabetic retinopathy has good anatomical results, increased postoperative vision, limiting bleeding and tears in surgery, limiting recurrent bleeding after surgery In Vietnam, the author Nguyen Thi Nhat Chau reports the results of using Bevacizumab in combination with vitrectomy for treatment of diabetic retinopathy, but the study was conducted on a small scale with short follow-up time, didn’t analyse related factors to the outcome 1.3.3.Related factors to the outcome of treatment 1.3.3.1.Systemic factors:age, duration of diabetes, high blood sugar, hypertension, diabetic kidney disease 1.3.3.2.Ocular factors related to treatment outcome * Diagnosis: vitreous hemorrhage gain the best result after treatment, vitreous hemorrhage accompanied by fibrosis proliferation also has good result if there are no complications or well-treated surgical procedures Vitreous hemorrhage with retinal detachment have the worst prognosis results *Complications: complications during surgery (bleeding, retinal tear, ), postoperative complications ( vitreous hemorrhage, iris neovascular, retinal detachment, cataract, infection .) * Additional treatment - Additional injections: recurrent vitreous hemorrhage, neovascular glaucoma - Additional surgery: additional surgery affects the results of anatomy, visual acuity and visual quality 10 CHAPTER SUBJECTS AND METHODS 2.1.Subjects: The study was conducted on severe proliferative diabetic retinopathy patients with vitreous hemorrhage complications who were examed and treated Retinal Department of National Institute of Ophthalmology, from January 1, 2012 to December 30, 2016 2.1.1.Selection criteria - Patients aged 18 years or older - Diagnosed with proliferative diabetic retinopathy with vitreous hemorrhage: + Vitreous hemorrhage does not clarify the fundus (grade II, III) + Vitreous hemorrhage with fibrosis proliferation or vitreous hemorrhage with retinal detachment of macula -Agree to join the research 2.1.2.Exclusion criteria - Patient has a history of vitrectomy - Patients currently have other eye diseases such as trauma, progressive inflammation, new glaucoma, etc - Patients with severe systemic diseases such as systemic disease, tuberculosis - If after intravitreal Bevacizumab injection, hemorrhage is fully absorbed or there is small amount of hemorrhage (grade I) without fibrosis proliferation/ retinal detachment patients move to laser 2.2.Methods 2.2.1.Study design: Intervention study had no control group 2.2.2.Sample size : n=68 2.2.3.Sampling method All patients eligible for study were selected up to the sufficient number 2.2.4.Research tools 2.2.4.1.Examination tools 2.2.4.2.Surgical tools: Surgical instruments and machines 2.2.5.Realization steps 2.2.5.1.Medical history records 13 No increased visual acuity after treatment: visual acuity stays the same or worse than pre-treatment visual acuity In the case of visual acuity at counting fingers level, if the visual acuity is still between ≤ meter between assessments, the visual acuity will not increase - Evaluating postoperative surgery results + Success: complete clear environment of the components of the fundus, no neovascular and fibrous membranes on the retina, optic disc, retinal attachment, retinal neovascularization does not progress further + Failure: opaque vitreous obscures the central retina or the entire retina, the fundus is not illuminated or retinal detachment, there are persistent retinal fluid shrinkage, neovascular continue to evolve, and neovascular glaucoma -Complications after surgery: recurrent vitreous hemorrhage, uveitis, endophthalmitis, cataract, retinal detachment, ocular atrophy, and other complications * Additional treatment: additional injections, additional surgery * General evaluation * Successful surgery - Anatomical results: remove blood in the vitreous chamber, remove all the hyaloids later, remove the fibrous proliferating membrane and contract - Vision increased at the time of examination compared with vision before treatment * Surgery failure: when either of the criteria or both criteria on vision and surgery are missing 2.2.6.3 Evaluate factors related to treatment results * Systemic factors associated with the outcome of treatment * Ocular factors associated with the outcome of treatment: diagnosis, complications, additional treatment 2.2.7 Data processing: SPSS 18.0 statistical software Tested by Tstudent algorithm and when squared, test Phi 2.2.8 Research ethics 14 CHAPTER RESULTS 3.1.PATIENT CHARACTERISTICS IN PRE-TREATMENT A total of 68 eyes that were studied, considering each eye to be an individual with independent systemic characteristics 3.1.1 Systemic characteristics: The mean age was 57.3 ± 8.4 years, in which aged under 64 years accounted for 82.4% The percentage of males and females was 58.8% and 41.2%, respectively Mostly type diabetes was identified in this study The mean duration of diabetes was 12,2 ± 6,8 years, in which the duration that under 15 and over 15 years were 60.3% and 39.7%, respectively Unstable treatment of diabetes was 97.1%, in which the rate of using Insulin was 75% Besides, the incidence of hypertension and diabetic nephropathy was 83.8% and 29.4%, respectively 3.1.2 Ocular characteristics 3.1.2.1 Related characteristics: The mean duration of diabetic retinopathy was 9,4 ± 11,7 months Patients treated with laser therapy before being included in this study was 16.2% and no laser therapy was 83.8% 3.1.2.2 Clinical findings: * Visual acuity: The mean visual acuity was 1.52±0.34 (logMar) (Counting finger meters) Poor visual acuity (≤ 20/400) accounted for 85.3%, while visual acuity ranged from 20/200 to 20/80 was 14.7%, no good visual acuity (≥ 20/63) in this study * Intraocular pressure:Normal *Diagnosis characteristics: 100% eyes with vitreous hemorrhage, of these, 77.9% were vitreous hemorrhage and 22,1% acootractional retinal detachment Grade I, II, III vitreous hemorrhage was accounted for 13.2%, 35.3%, 51.5%, respectively, showed significantly statistical difference * Posterior vitreous detachment characteristics: Complete, incomplete and no posterior vitreous detachment was 4.4%, 48.5%, 47.1%, respectively Posterior vitreous detachment in eyes with vitreous hemorrhage (45.3%) was significantly statistical difference, compared 15 with eyes with vitreous hemorrhage and tractional retinal detachment (80%) * Fibrovascular membrane characteristics: Fibrovascular membrane was reported 47.1%, no fibrovascular membrane was reported in 52.9% of cases * Lens characteristics: Cataract was found in 91.2%, no cataract was 4.4%, intraocular ocular lenses was 4,4% of cases, respectively 3.2 TREATMENT OUTCOME 3.2.1 Evaluation after intravitreal injection * Preoperative injection timing: range 5-14 days, in which 20 eyes had preoperative injection accounted for 29.4% (range, 10-14 days) * Mean of visual acuity: After intraocular injection, mean of visual acuity was 1.5 ± 0.39 ( logMar) ( approximately counting finger 2meters) * Intraocular pressure: 100% normal *Complications: There were 66 eyes (97.1%) without complications and cases (2.9%) with subconjunctival hemorrhage 3.2.2 Surgical evaluation 3.2.2.1 Surgical indication - There were 53 eyes (77.9%) with grade I,II,III vitreous hemorrhage, in which 18 cases with fibrovascular proliferation There were eyes diagnosed Grade I vitreous hemorrhage with fibrovascular proliferation 19 eyes diagnosed Grade I vitreous hemorrhage of which eyes with fibrovascular proliferation 31 eyes with Grade III vitreous hemorrhage of which eyes with fibrovascular proliferation - In case of vitreous hemorrhage with tractional retinal detachment, we documented 15 eyes (22.1%) in our study 3.2.2.2 Intraoperative evaluation * Surgical technique: 91.2% En-bloc technique, 8.8% combined technique * Intraocular tamponade: Using fluid, air, gas and silicon oil accounted for 61.7%, 16.2%,16.2%, and 5.9%, respectively * Cataract surgery:Combined cataract surgery with vitrectomy was reported in 58 eyes (85.3%) 16 * Intraoperative complications: No complications (58.8%), hemorrhage without retinal tear (23.5%), retinal tear ( 17.7%) 3.2.2.3 Post-operative evaluation * Visual acuity - The mean visual acuity at 1-week, 1-month, 3-month, 6-month postoperative follow-up was 1.26 ± 0.42, 0.98 ± 0.52, 0.78 ± 0.51, 0.75 ± 0.53 LogMar, respectively At 12-month postoperative follow-up in 67 eyes, the mean visual acuity was 0.74± 0.53.At 24-month postoperative follow-up in 66 eyes, the mean visual acuity was 0.78± 0.53 The visual acuity at 3, 6, 12, 24 months postoperatively was no significantly different (p>0.05) - The visual acuity was divided into groups, in which the initial poor vision in this study was 85.3% (58 eyes) Visual acuity was not improved after having Bevacizumab intravitreal injection 1-week postoperative , the good visual acuity was documented in eyes (5.9%), while most of patients had poor visual acuity 42 eyes (61.8%) At 1-month, 3-month, 6month postoperative the visual acuity was improved in 13 eyes (19.1%), 31 eyes (45.6%), 34 eyes (50%), respectively At 12 months and year follow-up, good visual acuity was determined in 30/67 eyes (44.8%) and 31/66 eyes (47%), respectively In group of moderate vision acuity, there was an improvement identified at preoperative, postoperative and a longterm follow-up In visually impaired group, there was markedly decreased from postoperative to the time of months follow-up (poor visual acuity was documented in ¼ patients) - The incidence of improved visual acuity after surgery 1-week (57.4%), 1-month (72.1%), 3-month (85.2%), 6-month (80.9%), 12month (80.6%) , 24-month (80,3%) In a group of moderate and good visual acuity, the incidence of increased visual acuity postoperatively that compared with pretreatment was significantly different (p 0.05) * Correlation between diagnosis and visual acuity group The visual acuity of vitreous hemorrhage group and vitreous hemorrhage with retinal detachment group at preoperative, postinjection evaluation was mostly poor vision group After surgery, good visual outcome in vitreous hemorrhage group was higher than vitreous hemorrhage with retinal detachment group (p< 0.05) *Correlation between diagnosis and anatomical outcome: The successful anatomical outcome of vitreous hemorrhage group (80.4%) showed no difference compared with vitreous hemorrhage with retinal detachment group (80%) (p>0.05) 3.3.2.2 Complications * Intraoperative complications: - Fibrovascular membrane is a factor affected to intraoperative complications - The rate of improved visual acuity, good, moderate, poor visual acuity and poor anatomical outcome in a group of no complications showed no significantly statistical difference, comparing with a group of complications (p > 0.05) * Postoperative complications: - Correlation between postoperative complications and improved vision: The rate of improved vision in a group of no complications (93.9%) showed no significantly statistical difference, comparing with a group of complications (66.7%) (p > 0.05) 19 - Correlation between postoperative complications and visual acuity group: The rate of good visual acuity in a group of no postoperative complication (66.7%) presented significantly statistical difference, comparing with a group of complications (27.3%) (p>0.05) - Correlation between postoperative complication and anatomical outcome: A group of no postoperative complication presented with 100% successful anatomic outcomes In a group of complication, anatomic outcome was achieved successfully in 60.6% cases and failed in 39.4% cases - Correlation between postoperative macular edema and improved visual acuity: The rate of postoperative group without macular edema (72.7%) presented no statistical difference, comparing with a group of macular edema (84.1%) 3.3.2.3 Additional treatment *Additional intravitreal injection: As the improved visual acuity at 24-month follow-up compared with pretreatment, visual acuity and anatomical outcome showed no significantly statistical difference between two groups of additional injection and non-additional injection * Additional surgery - Correlation between additional surgery and improved visual acuity: As the improved visual acuity at 24-month follow-up compared with pretreatment, there was significantly statistical difference between two groups of additional surgery (44.4%) and no additional surgery (93.8%) - Correlation between additional surgery and visual acuity: The result of visual acuity in a group of no additional surgery (62.5%) was significantly statistical difference, comparing with a group of additional surgery (5.6%) - Correlation between additional surgery and anatomical outcome: The rate of successful anatomy in a group of no additional surgery (89.6%) was significantly statistical difference, comparing with a group of additional surgery (55.6%) 20 CHAPTER DISCUSSION 4.1 PATIENT CHARACTERISTICS BEFORE TREATMENT 4.1.1 General characteristics: The mean age was 57.3 ± 8.4, under 64 years old accounted for 82.4% There was no significant difference in sex ratios, corresponding to the results of Bandello, El-Batany… Our study is mainly on type diabetes patients The mean duration of diabetes was 12.2 ± 6.8 years, in which the duration was under 15 years accounted for 60.3%, the rate of using Insulin was 75%, ineffective glycemic control was 97.1%, hypertension rate was 83.8%, kidney disease was 29/4%, resulting in Diabetic Retinopathy 4.1.2 Ocular characteristics 4.1.2 Ocular manifestations 4.1.2.1 Related characteristics: The mean time of diabetic retinopathy course was 9.4 ± 11.7 months, including 16.2% eyes with preoperative incomplete laser treatment and 83.8% eyes with no treatment The longer time of diabetic retinopathy course, the higher chance of getting worse, particularly, if diabetes is left untreated or improperly treated 4.1.2.2 Clinical ocular manifestations: The mean of postoperative visual acuity was 1.52±0.34 (logMar), primarly poor vision In our study, the percentage of normal IOP was 100%, 77.9% eyes with vitreous hemorrhage, 22.1% eyes with vitreous hemorrhage and tractional retinal detachment Vitreous hemorrhage, proliferative fibrotic membranes or tractional retinal detachment are causes leading to visual impairment - The percentage of incomplete posterior vitreous detachment and no posterior vitreous detachment was 48.5%, and 47.1%, respectively The progression of proliferation related to posterior vitreous detachment leading to an increased force in the adhesion area of the vitreous to the retina causes several complications, such as vitreous hemorrhage, and tractional retinal detachment The percentage of posterior vitreous detachment in vitreous hemorrhage group was 45.3% The difference 21 was statistically significant, compared to vitreous hemorrhage and tractional retinal detachment group accounted for 80% of cases - Fibrovascular membrane was found in 47.1% of cases Neovasculazation and fibrous tissue primarly develop along the surface of posterior hyaloid membrane With regards to posterior vitreous detachment, fibrovascular membranes further develop in vitreous cavity, leading to an increase of tractional force at the vitreoretinal adhesion, causing complications such as vitreous hemorrhage and retinal detachment - Older participants in this study were diagnosed with diabetic retinopathy, therefore, most of research eyes were diagnosed with cataracts (91.2%) 4.2 TREATMENT RESULTS 4.2.1 Evaluation after intravitreal injection - Preoperative intravitreal injection of Bevacizumab was accounted from to 14 days The proportion of eyes required surgical therapy after having intravitreal injection from 10 to 14 days due to the high blood surgar level was 29.4% - The mean visual acuity after intravitreal injection was 1.5 ± 0.39 logMar After having intravitreal injection, treated eyes still suffered dense vitreous hemorrhage In case of eyes with mild vitreous hemorrhage, we documented there was a proliferative fibrovascular membrane as well as tractional retinal detachment causing visual impairment - Intraocular pressure after injection was completely normal for all eyes This study only injected slowly 0.05 ml so that there was no elevated IOP in follow-up visits - Complications: no systemic complications because the participants with myocardial infarction or history of stroke were excluded in this study The incidence of patients in whom subconjunctival hemorrhage was 2.94% 4.2.2 Surgical evaluation 22 4.2.2.1 Surgical indication: After Bevacizumab intravitreal injection, there were eyes with grade I vitreous hemorrhage and proliferative fibrovascular, 19 eyes with grade II vitreous hemorrhage, 31 eyes with grade III vitreous hemorrhage, 15 eyes with vitreous hemorrhage and retinal detachment that required surgical intervention The indication of vitrectomy is to eliminate hemorrhage as well as traction to make retina reattach 4.2.2.2 Intraoperative evaluation: * Surgical technique: 62 eyes (91.2%) required En-bloc technique, eyes (8.8%) needed combined technique The complexity of vitrectomy for diabetic retinopathy depends greatly on the condition of the adhesive vitreoretina * Intraocular tamponade: The selection of intraocular tamponade based on preoperative lesions and intraoperative complications Silicon oil was used in 5.9% of cases due to complications, such as retinal tear, retinal tear combined with intraoperatively intensive hemorrhage that prevent performing laser therapy * Cataract surgery: The percentage of eyes combined with cataract surgery was 85.3% Regarding vitrectomy combined phacoemulsification and IOL implantation, the view of operative field for surgeons will be better and easier to perform clear vitrectomy, manage fibrovascular membrane, retinal lesions and laser therapy for peripheral retina * Intraoperative complications:Hemorrhage was found in 23.5% of cases, while retinal tear was identified in 17.7% of cases Retinal tear in this study is lower than N.N.Chau’s study (27%) The reason is that we performed Bevacizumab intravitreal injection before surgery, therefore, fibrovascular membrane was limited progressively as well as the removal of this membrane was easier, being effective in the prevention of intraocular complications 4.2.2.3 Postoperative evaluation * Visual acuity outcome: 23 - The mean visual acuity of postoperation was better (LogMar) Vision is relatively stable from 3-month follow-up At 24-month followup, the mean visual acuity was 0.78 ± 0.53 - Visual acuity group: Visual improvement was achieved gradually Good vision occured 1-week postoperation, greatly improved and being stable from 3-month follow-up Moderate and good vision in our study (74.3%) and Khan’s study (78%) are higher than N.N.Chau’s study (50.9%) The reason is that our study and Khan’s study performed Bevacizumab intravitreal injection before surgery, therefore, intraoperative and postoperative complications are limited - The rate of postoperative improved vision is gradually increased The percentage of increased visual acuity was 80.3%, similared to Khan’s study (78%) The visual acuity of a group of moderate-good vision was higher than a group of no improvement * Anatomical outcome: The result of successfully anatomical outcome in our study was 80.3%, being higher than NN.Chau’s study (72.2%) The reason is that we performed Bevacizumab intravitreal injection before surgery, making neovascular degenerated in quality as well as aperture Therefore, the dissection of fibrous membranes become easier and managing lesions well, leading to fewer intraoperative and postoperative complications * Postoperative complications: 26 eyes had postoperative vitreous hemorrhage (38.2%) Bevacizumab preoperative intravitreal injection help manage retinal lesions, reduces intraoperative complications and limits postoperative hemorrhage Most of postoperative hemorrhage was mild and be diminished by medical treatment The percentage of retinal detachment was 4.4% and visual acuity was not improved by vitrectomy Neovascular glaucoma was a severe complication and not responded with treatment Other complications: cataract, epiretinal membrane, high IOP, choroidal detachment, require additional surgical intervention 24 - Regarding macular edema, apart from the great role of tractional factor, edema would be remarkably reduced after 2-week follow-up of vitrectomy Therefore, other types of edema are not presumptive as complications or pre-existing lesions of diabetic retinopathy However, we reported that ocular manifestation has only been found after surgery 4.2.2.4 Additional treatment * Additional injection:Bevacizumab additional intravitreal injection is indicated for macular edema, recurrent postoperative vitreous hemorrhage, resulting in improved visual acuity Bevacizumab’s effect would be benefit in reducing the pathogenesis of edema and increasing blood absorption in vitreous cavity to limit additional vitrectomy The injection of Bevacizumab in neovascular glaucoma is to degenerate neovasular in retina and iris and lower IOP, preventing hemorrhage but the results showed no positive * Additional surgery:A total of eyes required the secondary, third and fourth surgery was 14.7%, 8.8% and 2.9% of cases, respectively There is a variety of additional surgery, such as retinal detachment, recurrent vitreous hemorrhage, neovascular glaucoma, cataract, epiretinal membrane, high IOP and choroidal detachment 4.2.2.5 General evaluation of treatment outcomes: Treatment outcome was successful in 74.2% of cases when Successful treatment outcome was 74.2% of cases when the success of visual acuity and anatomy were achieved at the time of evaluation 4.3 FACTORS RELATED TO TREATMENT OUTCOMES At 24 months follow-up, we examined 66 eyes in the study Although there are many factors related to treatment outcomes, we take into account the visual and anatomical outcomes 4.3.1 Systemic factors related to treatment outcomes: Age, sex, and duration of diabetes were not related to the results of treatment No association was identified between blood sugar level control, use of insulin, hypertension, kidney disease and treatment outcomes due to this 25 study has been conducted to a limited data, and whole body condition of patients was severe at the time of presentation 4.3.2.Ocular manifestation related to treatment outcomes 4.3.2.1 Diagnosis: There is no difference of improved visual and anatomical outcomes between a group of vitreous hemorrhage and vitreous hemorrhage with retinal detachment The effect of the treatment improves the quality of vision, postoperative recovery of visual acuity is gradually progresses to a good vision Good vision results in a vitreous hemorrhage group is higher than a group of vitreous hemorrhage with retinal detachment (p0.05) *Additional surgery: Additional surgery would affect considerably to the success of anatomical outcomes as well as visual function In a group of required additional surgery, poor anatomical outcomes determine poor postoperative rates However, not all success cases of anatomical outcome results in good visual acuity Proliferative diabetic retinopathy has several severe complications, and complicated pathogenesis causes retinal function Thus, although well anatomical outcome managed, the visual function was not improved 27 CONCLUSION The results of Bevacizumab (Avastin) intravitreal injection combined with vitrectomy to treat hemorrhage complications in proliferative diabetic retinopathy There were eyes with grade I vitreous hemorrhage and proliferative fibrovascular, 19 eyes with grade II vitreous hemorrhage, 31 eyes with grade III vitreous hemorrhage and 15 eyes with vitreous hemorrhage and retinal detachmentthat required surgical intervention The mean postoperative visual acuity was 0.78 ± 0.53 (logMar) The rate of improved vision and successfully anatomical outcome were 80.3% The rate of successful surgery was 74.2% of cases Hemorrhage complication during surgery was 23.5% and retinal tear was 17.7% of cases Regarding postoperative complications, vitrous hemorrhage was 38.2%, retinal detachment was 4.4%, neovascular glaucoma was also 4.4%, cataract was 3% complications could be combined as well Additional treatment, such as additional injection for recurrent vitreous hemorrhage in 10 eyes, neovascular glaucoma in eyes Injection to treat macular edema is in 32 eyes Additional surgery was indicated for 18 eyes Analysis some factors related to treatment outcomes of this method Age, sex, and duration of diabetes, status of diabetes, use of insulin, hypertension, kidney diseasewere not related to the results of treatment Preoperative diagnosis was not contributed to treatment outcomes The rate of good vision after treatment in a group of vitreous hemorrhage was higher than a group of vitreous hemorrhage with retinal detachment (p