Hematologic abnormalities associated with various types of retinal vein occlusion.

BACKGROUND: The objective of this study was two-fold: (1) to investigate hematologic abnormalities associated with various types of retinal vein occlusion (RVO) and comparison of their prevalence among those various types of RVO; (2) to review the conflicting literature on the subject, to place the information in perspective.
METHODS: In patients with various types of RVO seen in our clinic since 1973, we conducted planned prospective studies on the prevalence of: (1) routine hematologic tests (535 patients) and (2) certain special hematologic parameters (platelet aggregation, antithrombin III, and proportional, variant(2) globulin in 110, 81 and 91 patients, respectively). Patients were categorized into six types of RVO, based on defined criteria: non-ischemic and ischemic central RVO (CRVO), non-ischemic and ischemic hemi-CRVO (HCRVO), and major and macular branch RVO (BRVO). The patients had a detailed ophthalmic, systemic and hematologic evaluation. The data were abstracted and analyzed retrospectively from the detailed information originally collected prospectively in the patients' records. For data analysis, patients were divided into young, middle-aged and elderly. Observed prevalence rates of hematologic abnormalities were estimated. Logistic regression, adjusting for age and gender, was used to compare the observed prevalence of hematologic abnormalities among the various types of RVO.
RESULTS: No generalizations about the prevalence of hematologic disorders in all six types of RVO are possible. Ischemic CRVO showed a significantly higher prevalence of abnormal hematocrit ( P=0.044), hemoglobin ( P=0.018), and blood urea nitrogen ( P=0.025) than non-ischemic CRVO, while a significantly higher prevalence of abnormal antinuclear antibody (ANA; P=0.049) was seen in non-ischemic CRVO than in ischemic CRVO. There was a significant ( P=0.011) difference in the prevalence of abnormal uric acid among the three main RVO groups (CRVO, HCRVO, BRVO), highest in BRVO and lowest in HCRVO. There was a higher prevalence of abnormal glucose ( P=0.069) and ANA ( P=0.071) in CRVO+HCRVO than in BRVO. Results of special hematologic studies are given.
CONCLUSIONS: Our study showed that a variety of hematologic abnormalities may be seen in association with different types of RVO, and any generalization about these disorders applied to all RVO patients may be misleading. The evidence of our study and in the literature indicates that there is no good reason why all patients with RVO should be subjected to extensive, expensive, special hematologic and hypercoagulability investigations, unless, of course, there is some clear indication; the routine, inexpensive hematologic evaluation is usually sufficient for RVO patients. Treatment with anticoagulants or platelet anti-aggregating agents may adversely influence the visual outcome, without any evidence of protective or beneficial effect.