In this article, the pathophysiological and clinical relevance of the evaluation of ocular surface temperature (OST) in the management of central retinal vein occlusion (CRVO) will be discussed. First, an updated overview of the disease will be provided in order to establish a scientific background for the link between CRVO and OST. Second, the actual role of the measurement of OST in ophthalmology when the reading is obtained by means of infrared thermography will be clarified. Third, the technical features of this device and methods of examination will be explained. Finally, literature data about the object of this article will be presented and future perspectives of employment of OST evaluation in the work-up of patients affected with CRVO will be discussed.
Central Retinal Vein Occlusion – An Update
CRVO is a common sight-threatening vascular disorder. The estimated prevalence varies from 0.4 to 1%, and the visual prognosis is severe in at least 50% of patients.1–3 The aetiopathogenesis of CRVO is multifactorial, involving both systemic and local factors. It has been ascertained that traditional atherosclerotic risk factors, such as arterial hypertension, diabetes, dyslipidaemia and smoking, are more prevalent in patients with CRVO than in healthy controls.4–7 The association of CRVO with thrombophilia, hypofibrinolysis and blood hyperviscosity has been demonstrated.8–11 Local factors related to an increased risk of developing CRVO include glaucoma and inherited or acquired features of lamina cribrosa predisposing to venous stasis – and in general to ocular blood flow (OBF) alteration. Pathogenetically speaking, CRVO occurs when a permanent thrombosis is present in the central retinal vein, producing the irreversible occlusion of the vessel, which leads to the typical clinical symptoms and signs of the disease.4,12,13
The classification of CRVO distinguishes between ischaemic CRVO and non-ischaemic CRVO.12–15 In the majority of cases, ischaemia is not present at the time of diagnosis, but becomes evident during follow-up. The time between CRVO onset and eventual ischaemic CRVO development is variable, with the highest incidence in the first six months. Considering all CRVOs, the cumulative incidence of the ischaemic type, which leads to retinal, disc and iris neovascularisation, is about 20%, while neovascular glaucoma occurs in 9–10% of cases. These data refer to observations performed at three to six months after the diagnosis of the disease. Considering only ischaemic CRVOs, the incidence of neovascular glaucoma increases to 40% after the same period of follow-up.13
Differentiation between non-ischaemic and ischaemic CRVO is crucial for the management of the disease. This is because retinal ischaemia, through stimulating the production of vascular endothelial growth factors (VEGFs), leads to complications such as retinal vascularisation and neovascular glaucoma, which are responsible for severe visual loss and pain in CRVO eyes. Non-ischaemic CRVO is a relatively benign disease, with a central scotoma resulting from macular oedema as the only complication.12–15 Furthermore, laser grid photocoagulation and corticosteroid intravitreal injections revealed their efficacy in reducing macular oedema and then improving visual function. Conversely, no treatment proved to be effective in preventing the development of retinal ischaemia.
Therapeutic strategies available to manage ischaemic CRVO – including laser photocoagulation, anti-VEGF intravitreal injections and radial optic neurotomy – cannot improve the poor visual prognosis that characterises the ‘complicated’ CRVOs. Currently, only regular, strict clinical observation is recommended as early as possible to detect progression to ischaemic CRVO.13,16–19 The goal of the management of CRVO patients is identifying in the acute phase those eyes that are susceptible to developing the ischaemic type.
Fluorescein fundus angiography (FFA) is considered the gold standard to diagnose the conversion of non-ischaemic CRVO to the ischaemic type.14 Unfortunately, it allows ischaemic damage to the retina that is already present to be ascertained. It is important to be able to predict the development of ischaemia from the first days after the onset of CRVO.13 Additional tests to achieve this were proposed by Hayreh et al. in 1990.20 They included visual and acuity field tests, a relative afferen papillary defect test and an electro-retinogram. In the early acute phase they proved to be more reliable than FFA, especially when considered all together. Nevertheless, these functional tests have not replaced FFA, mainly because the guidelines for the treatment of CRVO rely on the conclusions of the CRVO Study. According to the authors, “10-disc area of retinal capillary obliteration” on FFA is the gold standard to differentiate the two types of CRVO.
In this trial, pan-retinal laser photocoagulation (PRP) was performed in eyes that developed two clock-hours of iris neovascularisation or any angle neovascularisation.14 Although PRP has been shown to be effective in preventing neovascular glaucoma, the identification of ischaemic CRVOs at an earlier stage remains a concern that deserves to be considered to ameliorate the clinical outcome of patients affected with CRVO.
During the last few years, the interest of some authors has been pointed at detecting an objective, non-invasive and reliable test for the differentiation of the two types of CRVO. Based on the pathophysiology of the disease, it is evident that the evaluation of ocular blood flow (OBF) features, particularly ocular haemodynamics, could provide relevant information that would be potentially useful in the follow-up of CRVOs, aiming at identifying those eyes that will develop severe ischaemia. In clinical practice, three diagnostic devices have been tested: ophthalmodynamometry, colour Doppler imaging (CDI) and infrared thermography.
An ophthalmodynamometric study showed that in acute CRVO, diastolic central retinal vein pressure was significantly higher in eyes with ischaemic CRVO than in eyes with non-ischaemic type and in controls; moreover, in the ischaemic type the central retinal vein pressure was higher than the diastolic retinal artery pressure.21 Arséne et al. conducted a study by means of CDI to evaluate haemodynamic differences between non-ischaemic and ischaemic CRVOs during the first year after the diagnosis. They found a persistent impairment of central venous flow in ischaemic eyes, displayed by significantly lower blood flow velocities. Another finding of this trial was a more marked increase in the resistivity index of the central retinal artery in ischaemic CRVO eyes at presentation.22
The clinical relevance of the two studies above relies on the simplicity, non-invasiveness and reliability of the methods, although standardisation is needed. These results suggest that the methods of examination discussed above could have a prognostic value; this should be confirmed in larger populations.
Ocular Surface Temperature and Its Relationship with Central Retinal Vein Occlusion
OST values are determined by extraocular factors, such as body and environmental temperature, and ocular factors, such as the features of the tear film layer, inflammatory and neoplastic diseases and OBF.23 Previous investigations have demonstrated that changes in OST may occur in the presence of omolateral carotid artery stenosis, conjunctivitis, anterior uveitis, ocular neoplasms and glaucoma, and following cataract and refractive surgery.24–27 Some authors have indicated a possible correlation between OST and OBF, in particular with retrobulbar haemodynamics assessed by means of CDI.27,28 A correlation between OST and OBF has been clearly demonstrated in healthy subjects. Of note, lower OST values were associated with worse retrobulbar CDI parameters in glaucoma eyes.
Infrared thermography is a simple, reliable, non-invasive and rapid device that measures the radiated heat from the body surface. It is widely used to study abnormalities in blood flow in rheumatic diseases, dentistry, oncology and vascular disorders. This technique is also suitable for measuring ocular surface temperature, providing information about the features of local blood flow. Changes in ocular haemodynamics are crucial in the pathophysiology of CRVO; therefore, this technique could be helpful in detecting OBF abnormalities that may lead to the development of retinal ischaemia. Infrared thermography could provide evidence that eyes affected with CRVO-typical alterations of ocular circulation are recorded as changes in OST values. More importantly, the realtime measurements of OST could help in differentiating those eyes that will develop retinal ischaemia in the acute phase of the disease.
To test these hypotheses, we decided to conduct a clinical trial. The purpose was to evaluate whether thermographic analysis obtained by means of infrared thermography in patients with CRVO may be helpful in the management of the disease.29 This is the first report on the topic. In fact, to the best of our knowledge, no literature data concerning ocular thermographic patterns in eyes affected with CRVO are available.
Technical Features of Ocular Surface Temperature Measurements
Before discussing our results, we will provide some information about the methods used in our investigation. We followed the methods validated by previous authors.26
To achieve reliable thermographic analyses in patients affected with CRVO, certain conditions regarding the patients and the environment are required. Room temperature, humidity and air flow must be relatively constant. Subjects must be free of body temperature alterations, dry-eye syndromes and any systemic and ocular pathologies that can influence OST, as well as carotid artery stenosis, inflammation or neoplastic states of the adnexa and the eyeball and glaucoma. We also excluded patients who underwent cataract and/or refractive surgery. In our trial, we maintained a room temperature of 24–25°C and a humidity of 55%, and avoided air-flow variations by keeping doors and windows closed. We selected patients with a body temperature between 36.4 and 36.7°C, with a normal ophthalmological examination except for CRVO and Schirmer’s test >10mm for five minutes, a break-up time >10 seconds and normal CDI examination of the epiaortic vessels. To avoid bias due to an increase of OST throughout the day, OST measurements were obtained between 9 and 10am.
All of the exams were performed using an infrared detector (Agema Thermovision 800 LWB, AGEMA Infrared Systems 1991 AB, Donderyd, Sweden) after a 20-minute period of adaptation within the room. The thermocamera was placed in front of the patient, who was requested to keep both eyes closed for three to five seconds, then to open them wide. OST measurements lasted for 20 seconds, and the data were registered every second, but only the first photogram of the film was considered for subsequent analyses. The temperature of five anatomical points across a line running horizontally through the centre of the cornea was recorded: the internal canthus (point 1), half-way from the internal canthus and nasal limbus (point 2), the centre of the cornea (point 3), half-way from the temporal limbus and external canthus (point 4) and the external canthus (point 5). Of note, OST registered at the centre of the cornea, i.e. corneal surface temperature, is the most reliable measure because the avascularity of the tissue should minimise the influence of the conjunctival vascular network on thermographic analyses. Corneal surface temperature is the most reliable OST indicator of blood flow at the level of the posterior segment of the eye.
Literature Data and Future Perspectives on the Ocular Surface Temperature in Central Retinal Vein Occlusion
As stated above, a recent trial conducted by our group is the only contribution to the evaluation of a possible link between OST and CRVO. Following up this notion, we will present and discuss our findings. Our results showed that the OST values measured in CRVO eyes were lower than those in control eyes. Within the CRVO group, the affected eyes showed lower OST values at all five points in comparison with the non-affected ones. However, the statistical significance was achieved only at point 3, the centre of the cornea. Interestingly, OST readings in non-affected eyes of patients were higher than in CRVO ones, but lower than in controls. This finding could indicate a more general impairment of blood flow in an individual who experienced a thrombotic event, and low OST measures are likely to be interpreted as markers of the local blood stasis that occurs in CRVO.
In this study, thermography was performed within one month from the onset of the symptoms in order to have greater differentiation among the groups of studied eyes in terms of vascular alterations. We divided the CRVO eyes between non-ischaemic and ischaemic on the basis of the FFA performed one month after the onset of the disease, then we analysed the differences between OST values in the two subgroups. Even though a statistical analysis was not feasible because of the small sample size, we reported a trend indicating lower OST values in ischaemic CRVO eyes compared with ischaemic ones. This finding suggested a possible prognostic value for ocular thermography in eyes with acute CRVO. In our opinion, the hypothesis that lower OST measures are associated with greater retinal ischaemic areas deserves to be tested in a larger population study in order to establish a possible prognostic value for ocular thermography.
In conclusion, ocular surface measurements seem to be suitable for evaluating blood-flow alterations in patients affected with CRVO in the early stages of the disease. Infrared thermography should be considered at least as an additional imaging method to be used in the clinical workup of CRVO patients.
