|Year : 2014 | Volume
| Issue : 4 | Page : 174-178
Preoperative prognostic factors in vitrectomy for severe proliferative diabetic retinopathy
Shan-Jiun Lin, Po-Ting Yeh, Jehn-Yu Huang, Chung-May Yang
Department of Ophthalmology, National University Hospital; Medical College, National University, Taipei, Taiwan
|Date of Web Publication||1-Oct-2014|
Department of Ophthalmology, National Taiwan University Hospital, College of Medicine, National Taiwan University, Number 7, Chung-Shan South Road, Taipei
Source of Support: None, Conflict of Interest: None
Purpose: To investigate the preoperative prognostic factors contributing to extremely poor visual outcome in cases of proliferative diabetic retinopathy after a successful vitrectomy and an uneventful postoperative course.
Methods: We enrolled 28 consecutive eyes of 28 patients with poor preoperative visual acuity (VA), who underwent primary diabetic vitrectomy. The postoperative course was uneventful, and retinal attachment was achieved for at least 6 months in all cases. The cases were separated into the study group (postoperative VA < preoperative VA) and control group (postoperative VA ≥ preoperative VA). Preop-erative factors including systemic diseases, demographic data, proliferative diabetic retinopathy severity, and neovascularization status were compared between the two groups. Significant risk factors for poor visual results were determined by logistic regression analysis.
Results: The creatinine level was significantly higher in the study group (4.07 ± 4.15 mg/dL) than in the control group (1.23 ± 0.46 mg/dL; p = 0.003). Chronic macular detachment was noted in 7/10 eyes (70%) in the study group and in 1/18 eyes (5.6%) in the control group (p = 0.001). Broad fibrovascular proliferation extending to the periphery for more than two quadrants was found in 4/10 eyes (40%) in the study group and in none of the eyes in the control group (p = 0.016). Chronic macular detachment and broad fibrovascular proliferation were significantly associated with poor visual outcome in multiple logistic regression analysis.
Conclusion: Preoperative chronic macular detachment, broad fibrovascular proliferation, and poor renal function may indicate worse visual results after successful diabetic vitrectomy for cases with severe diabetic retinopathy.
Keywords: creatinine fibrovascular proliferation macular detachment proliferative diabetic retinopathy vitrectomy
|How to cite this article:|
Lin SJ, Yeh PT, Huang JY, Yang CM. Preoperative prognostic factors in vitrectomy for severe proliferative diabetic retinopathy. Taiwan J Ophthalmol 2014;4:174-8
|How to cite this URL:|
Lin SJ, Yeh PT, Huang JY, Yang CM. Preoperative prognostic factors in vitrectomy for severe proliferative diabetic retinopathy. Taiwan J Ophthalmol [serial online] 2014 [cited 2019 Nov 12];4:174-8. Available from: http://www.e-tjo.org/text.asp?2014/4/4/174/204135
| 1. Introduction|| |
Vitrectomy is the major treatment modality for severe pro-liferative diabetic retinopathy (PDR). With advances in both the understanding of pathoanatomy and vitrectomy techniques and instruments, a high anatomical success rate can be achieved in eyes with severe PDR. However, improvement of visual outcome might not occur in all cases. Various pre- and postoperative factors associated with a poor visual outcome have been reported in the literature, including poor preoperative visual acuity (VA), older age, iris neovascularization,,, ahistory of lensectomy, the creation of iatrogenic breaks, and elevated intraocular pressure., Although these studies provide important information regarding the association between the periop-erative conditions and the anatomical and functional outcomes, few studies have specifically focused on the preoperative systemic and local factors that may be related to an extremely poor visual function despite an anatomical success and an uneventful postoperative course. In this study, we selected patients with poor initial vision who had obtained a successful anatomical outcome after an uneventful surgery and who experienced a smooth postoperative course. Cases with an extremely poor functional outcome (VA less than counting fingers) were compared with those that had a better visual outcome. Preop-erative systemic and local factors were examined to identify possible preoperative risk factors associated with a poor visual outcome in severe PDR. The identification of these factors may be useful for predicting poor functional outcome and avoiding unnecessary surgical interventions.
| 2. Patients and methods|| |
From January 2009 to March 2012, the clinical records of consecutive patients with poor preoperative VA (less than counting fingers) who underwent primary pars plana vitrectomy for complications of PDR were reviewed retrospectively. All cases were postoperatively followed up for >6 months. Patients with retinal attachment at the end of the surgery, an uneventful postoperative course, regular postoperative follow-up, and postoperative retinal attachment for at least 6 months were enrolled. Patients who had elevated intraocular pressure of > 35 mmHg for 2 consecutive postoperative days and patients who had significant vitreous hemorrhage and/or dense cataract before the surgery were excluded. Thus, only those cases whose visual function reflected the severity of the anatomical changes were selected. The study cases were separated into two groups according to the postoperative VA. We defined the study group as cases with a postoperative VA that was worse than the preoperative VA (hand motion, light perception, or no light perception) and the control group as cases with a postoperative VA better than the preoperative VA (counting fingers or any Snellen acuity). The postoperative VA was measured at least 6 months postoperatively. This study was approved by the ethics board of the National Taiwan University Hospital, Taipei, Taiwan.
2.1. Surgical technique
All of the surgeries were performed by one surgeon (C.M.Y.) using a standard three-port pars plana vitrectomy, as described previously., Briefly, a 20-gauge vitrectomy system was set up in every case. Anterior–posterior traction release was attempted first, followed by fibrovascular tissue removal with delamination as the principal technique. Hemostasis was achieved by raising the infusion bottle and applying mechanical compression with a soft-tipped cannula, endodiathermy, or a combination of the above techniques. Fluid–air exchange with internal drainage of the sub-retinal fluid was performed through preexisting or iatrogenic breaks followed by supplementary panretinal photocoagulation extending beyond the equator and peripheral cryotherapy. Long-acting gas (15–20% C3F8) or silicone oil (5000 cs) infusion was performed if needed. In certain cases, a 360° encircling buckle was placed to counter any possible residual peripheral vitreous traction. Bevacizumab (0.05 mg) was injected through an upper sclerotomy with an insulin syringe in every case.
The demographic data of the patients and the preoperative and intraoperative factors associated poor visual outcome were recorded in detail. The investigated preoperative general factors included sex, age, and underlying systemic diseases as well as local factors, including lens status, the presence or absence of chronic long-standing macular detachment, the extent of fibrovascular proliferation, predominantly active or inactive neovascularization, the extent of retinal detachment, and the existence or lack of iris neovascularization.
Chronic long-standing macular detachment was defined as four-quadrant vessel sheathing, a pale disc, uniformly fibrotic proliferation, and atrophic thinning of the detached macula if optical coherence tomography data were available. A predominantly active neovascularization was defined as the presence of visible neo-vascularization in the larger part of the proliferative tissue with any degree of vitreous hemorrhage, and a predominantly inactive neovascularization was defined as a fundus status in which most of the proliferative tissue appeared avascular. The extent of fibrovas-cular proliferation were classified as follows: Grade 0, a lack of adhesion or only focal adhesion; Grade 1, a broad adhesion of less than or equal to three sites; Grade 2, a broad adhesion of more than three sites, without extension to the periphery; Grade 3, a broad adhesion of more than three sites that extended to the periphery for less than or equal to two quadrants; and Grade 4, a broad adhesion of more than three sites that extended to the periphery for more than two quadrants. The extent of retinal detachment was classified as follows: localized only to the macular area, outside of the arcade for ≤ 2 quadrants, or outside of the arcade for > 2 quadrants.
To calculate the VA, vision of counting fingers, hand motion, light perception, and no light perception were respectively defined as a Snellen chart of 0.005, 0.002, 0.001, and 0.0005 and were subsequently converted to 2.3, 2.7, 3.0, and 3.3 logMAR units.
2.2. Statistical analysis
To examine the significance of predisposing factors correlated with poor visual outcome between the study group and the control group, the Chi-square test or Fisher’s exact test was performed for categorical variables and the Mann–Whitney U test was used for continuous variables. Fundus characteristic variables that appeared to have a possible association with visual outcome were first assessed by univariate logistic regression analysis. Significant factors were entered into a model using multiple logistic regression with the forward conditional method to verify the correlations further. All statistical analyses were conducted with SPSS statistics 17.0 (SPSS Inc., Chicago, IL, USA). Ap value of < 0.05 was considered statistically significant.
| 3. Results|| |
We enrolled 10 eyes of 10 patients (8 women and 2 men) in the study group and 18 eyes of 18 patients (14 women and 4 men) in the control group. The demographic data, preoperative clinical findings, and functional outcomes of all patients are listed in [Table 1]. The preoperative VA of both groups was less than counting fingers, and no significant difference was found in terms of age and preoperative VA between the two groups (p = 0.11 and p = 0.25, respectively). The mean postoperative VAs of the study group and the control group were 2.88 ± 0.29 logMAR units and 1.55 ± 0.52 logMAR units, respectively (p < 0.001). The mean improvement of the VA was 0.12 ± 0.29 logMAR units in the control group and -0.16 ± 0.52 logMAR units in the study group (p < 0.001; [Table 2]). The clinical features of both groups are listed in [Table 3]. Regular hemodialysis was performed in 3 of 10 patients (30%) in the study group, but in none of the patients in the control group. Furthermore, the creatinine level was significantly higher in the study group (4.07 ± 4.15 mg/dL) than in the control group (1.23 ± 0.46 mg/dL; p = 0.003). Chronic long-standing macular detachment was noted in seven of 10 eyes (70%) in the study group, but in only one of 18 eyes (5.6%) in the control group (p = 0.001). Four of 10 eyes (40%) in the study group had fibrovascular proliferation extending to the periphery for more than two quadrants, whereas all of the eyes in the control group contained fibrovascular tissue extending to the periphery for less than two quadrants (p = 0.016). Combined traction and rhegmatogenous retinal detachment was found in seven of 10 eyes (70%) in the study group and in nine of 18 (50%) eyes in the control group (p = 0.43). Cataract surgery was performed in the same setting in five of 10 eyes (50%) in the study group and in two of 18 eyes (11.1%) in the control group (p = 0.063; [Table 3]). When fundus variables were applied as parameters, univariate analysis showed that chronic long-standing macular detachment, the extent of fibrovascular proliferation, and retinal fibrosis were significantly associated with a poor prognosis (p = 0.003, p = 0.018, and p = 0.014, respectively). Multiple logistic regression analysis indicated that chronic long-standing macular detachment (p = 0.023) and the extent of fibrovascular proliferation (p = 0.049) were the strongest risk factors for poor visual outcome [Table 4]. An example from the study group is shown in [Figure 1] and an example from the control group is shown in [Figure 2].
|Figure 1: (A) Fundus photograph of Patient 1 shows fibrovascular proliferation with broad adhesion of more than three sites that extend to the periphery for four quadrants (Grade 4), chronic macular detachment, and predominantly inactive neovascularization. (B) After the operation, the retina was attached but the visual acuity remained hand motion. (C) Optical coherence tomography shows atrophic thinning of the retina.|
Click here to view
|Figure 2: (A) Fundus photograph of Patient 25 shows fibrovascular proliferation with broad adhesion of more than three sites that extend to the periphery for two quadrants (Grade 3), and predominantly active neovascularization. (B) After the operation, the retina was attached, and visual acuity improved to 0.1 from hand motion. (C) Vertical section of the optical coherence tomography image shows attached retina without central thinning.|
Click here to view
|Table 2: Preoperative and postoperative logMAR visual acuity in study and control groups.|
Click here to view
|Table 4: Logistic regression analysis for postoperative poor vision and fundus characteristic.|
Click here to view
| 4. Discussion|| |
This study specifically focused on the preoperative factors that contribute to extremely poor visual outcome (VA less than counting fingers, i.e., hand motion, light perception, or no light perception) in eyes that have undergone successful surgery (100% reattachment with at least 6 months of follow-up after surgery). Although a poor preoperative VA is reportedly a negative predictive factor for vit-rectomy in severe proliferative diabetic patients, it is this group of patients who may benefit most from VA gain via surgery. Although all of the patients had a poor preoperative VA in our study (less than counting fingers), improved visual outcome was achieved in the patients in the control group. Thus, the purpose of this study was to identify other preoperative factors associated with extremely poor visual function after an uneventful intra- and postoperative course in this group of patients. Risk factors that have been described in the literature as well as specific retinal findings were examined in order to correlate them with the final visual outcome. The study group had a significantly higher creatinine level compared to the control group. In addition, univariate statistical analysis using fundus variables as risk factors revealed that chronic long-standing macular detachment, the extent of fibrovascular proliferation, and predominantly inactive neovascularization were significantly associated with a poor prognosis. Multivariate analysis further showed that chronic long-standing macular detachment and the extent of fibrovascular proliferation were the most important factors that determined the final poor visual outcome.
Both diabetic retinopathy and diabetic nephropathy are among diabetic microvascular complications. The microangiopathy in the kidney may indicate a similar microvascular environment in the retina. In addition, the presence of one preexisting complication (diabetic retinopathy, diabetic nephropathy, or diabetic neuropathy) may contribute to the development of another complication. Our data showed that the patient’s creatinine level and whether or not the patient was undergoing hemodialysis were significantly related to the outcome of surgery for PDR. We postulate that poor renal function reflects the general worsening of the microvascular environment. Surgery on a retina with extensively damaged microcirculation and preexisting poor visual function may result in a poor prognosis after surgery.
In our study, the extent of fibrovascular proliferation was also significantly associated with a poor prognosis. The most critical and difficult part of vitrectomy for severe PDR is the meticulous dissection and removal of the preretinal membrane, and a favorable anatomical result depends on the proper removal of the fibrovas-cular tissue. A fibrovascular proliferation extending to the periphery not only indicates a fulminant fibrovascular proliferation, which leads to a more distorted and detached macular, but also complicates the surgery and prolongs the surgical period. The level of fibrovascular proliferation was classified into five grades in our study, and we found that this level was significantly associated with visual outcome. None of the patients in the control group had fibrovascular proliferation extending beyond the equator for more than two quadrants, whereas this was observed in 40% of the patients in the study group.
In this study, chronic long-standing macular detachment was another factor correlated with poor visual outcome in our study. Although this result may not be surprising, our literature review did not find reports discussing this fundus changes as a risk factor for poor visual prognosis. In this study, we specifically defined chronic long-standing macular detachment as the presence of diffuse vascular sheathing, atrophic thinning of the retina, uniformly fibrotic proliferation, and atrophic thin retina on optical coherence tomography images. We found that fundus changes containing these defined features were indeed associated with poor visual outcome. The PDR is no longer active when it has progressed to the burned-out stage, and a reduction in the caliber of the retinal vessels and sheathed vessels are often found in this stage. The quiescent neovascular activity may also reflect atrophy of the retina. Thus, a more advanced fibrotic stage indicates a more prolonged compromise of retinal conditions, which might be a factor for a poor prognosis. Although predominantly inactive neo-vascularization alone was not significantly correlated with a poor prognosis in our study (p = 0.125), 60% of the patients in the study group had predominantly inactive neovascularization compared to only 28% of patients in the control group. As the PDR with macular detachment progresses to purely inactive fibrotic proliferation on the macula, it usually reaches the atrophic stage. An atrophic optic nerve can be a sign of chronic long-standing macular detachment or compromised disc circulation secondary to ischemia, vitre-opapillary traction, or unstable intraocular pressure. Although the evolution of sophisticated vitrectomy instruments permits salvage of many cases of severe PDR, chronic long-standing macular detachment remains one of the most important indicators of poor functional outcome. Our study suggests that the necessity of the surgery should be reconsidered when signs such as a pale disc, four-quadrant vessel sheathing, pure fibrotic proliferation, and atrophic thinning of the detached macula are observed.
Although previous studies have provided multiple predisposing factors associated with poor visual outcome in vitrectomy for PDR, this study specifically investigated the preoperative factors for the most severe cases to evaluate the necessity of performing the surgery. Patients who have signs of chronic long-standing macular detachment, fibrovascular proliferation extending beyond the equator for more than two quadrants, and poor renal function, especially those who are undergoing regular hemodialysis, might be unsuitable candidates for surgery. Although we have found that certain preoperative factors may be correlated with extremely poor visual outcome and make surgical intervention a futile effort, it is more important to know how each factor weighed as a determinant of the visual prognosis as well as if the presence of multiple risk factors further increases the possibility of extremely poor final vision. Unfortunately, the small case number in the study group prevented such an analysis.
Other limitations of our study included its retrospective nature and the fact that the difference between the surgical procedures was not considered between cases. Although study bias was inevitable, this was kept as minimal as possible by careful documentation of the preoperative findings, which were confirmed during surgery, and by excluding patients with any surgical complications during long-term follow-up. Our study had several strengths: we excluded moderately severe cases by restricting the study group only to those with extremely poor postoperative VA and excluded cases complicated by vitreous hemorrhage. In addition, we selected those whose age and preoperative VA matched the control group patients. With these restrictions, the preoperative fundus features and systemic problems that are correlated with extremely severe PDR could be identified. Further prospective studies with a larger number size are necessary to confirm our conclusion.
Conflicts of interest: The authors have no conflicts of interest to declare.
| References|| |
Machemer R, Buettner H, Norton EW, Parel JM. Vitrectomy: a pars plana approach. Trans Am Acad Ophthalmol Otolaryngol
La Heij EC, Tecim S, Kessels AG, Liem AT, Japing WJ, Hendrikse F. Clinical variables and their relation to visual outcome after vitrectomy in eyes with diabetic retinal traction detachment. Graefe’s Arch Clin Exp Ophthalmol
. 2004;242: 210–217.
Newman DK. Surgical management of the late complications of proliferative diabetic retinopathy. Eye. 2010;24:441–449.
Yang CM, Su PY, Yeh PT, Chen MS. Combined rhegmatogenous and traction retinal detachment in proliferative diabetic retinopathy: clinical manifestations and surgical outcome. Can J Ophthalmol
Thompson JT, de Bustros S, Michels RG, Rice TA. Results and prognostic factors in vitrectomy for diabetic traction retinal detachment of the macula. Arch Ophthalmol
Thompson JT, Auer CL, de Bustros S, Michels RG, Rice TA, Glaser BM. Prognostic indicators of success and failure in vitrectomy for diabetic retinopathy. Ophthalmology
Yang CM, Yeh PT, Yang CH. Intravitreal long-acting gas in the prevention of early postoperative vitreous hemorrhage in diabetic vitrectomy. Ophthalmology
Yeh PT, Cheng CK, Chen MS, Yang CH, Yang CM. Macular hole in proliferative diabetic retinopathy with fibrovascular proliferation. Retina
Eliott D, Lee MS, Abrams GW. Proliferative diabetic retinopathy: principles and techniques of surgical treatment. In: Ryan SL, Hinton DR, Schachat AP, Wilkinson CP, eds. Retina
. 4th ed.. Philadelphia: Elsevier Mosby; 2006: 2413–2449; vol. 3.
Yuzurihara D, Iijima H. Visual outcome in central retinal and branch retinal artery occlusion. Jpn J Ophthalmol
Girach A, Vignati L. Diabetic microvascular complications–can the presence of one predict the development of another? J Diabetes Complications
. 2006;20: 228–237.
Danis RP, Davis MD. Proliferative diabetic retinopathy. In: Duh E, ed. Diabetic Retinopathy
. Totowa, NJ: Humana Press; 2008:32–39.
Kroll P, Wiegand W, Schmidt J. Vitreopapillary traction in proliferative diabetic vitreoretinopathy [see comments]. Br J Ophthalmol
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4]