Graefes Arch Clin Exp Ophthalmol DOI 10.1007/s00417-015-3039-6
RETINAL DISORDERS
Retinal vessel oxygen saturation and vessel diameter in retinitis pigmentosa at various ages Yao Zong 1 & Leilei Lin 1 & Changxian Yi 1 & Xia Huang 1 & Yue Fu 1 & Yanmin Dong 1 & Xiaobing Qian 1 & Yujie Li 1 & Qianying Gao 1
Received: 11 December 2014 / Revised: 16 April 2015 / Accepted: 28 April 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study was conducted to determine whether oxygen saturation and retinal blood vessel diameter are affected by retinitis pigmentosa (RP) at various ages. Methods Relative oxygen saturation was measured in retinal blood vessels in 68 RP patients and 136 normal subjects using the Oxymap T1 retinal oximeter. Subjects were divided into two age groups: Group A (20–40 years) and Group B (> 40 years). One randomly selected eye of each subject was used for statistical analysis. Student's t tests were used to analyze the mean saturation and diameter of retinal arterioles and venules and arteriovenous differences between RP and normal subjects in the two age groups. A Spearman test was used to analyze the correlation of mean saturation of retinal arterioles (AS) and arteriovenous differences (AVS) with visual acuity, disease duration, and electroretinogram (ERG) b-wave amplitude in patients with RP. Results AS was significantly higher in patients with RP (105.5±9.4 %) than in normal subjects (94.5±4.4 %, p= 0.000) in Group A, while in Group B, AS was significantly lower in RP patients (86.8±10.3 %) than in healthy subjects (96.0±4.8 %, p=0.000). Vessel diameter was smaller in RP patients than in normal subjects. AS and AVS showed a negative correlation with disease duration and a tendency toward positive correlation with ERG b-wave in patients with RP. Conclusions The shifting characteristics of retinal vessel oxygen saturation suggest that the pathological mechanism of
* Qianying Gao [email protected] 1
State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University,, 54 Xianlie Road, Guangzhou, Guangdong 510060, China
retinal oxygen metabolic disorder differs by age in patients with RP. Keywords Hypoxia . Oximetry . Oxygen saturation . Retina . Retinitis pigmentosa
Introduction As an inherited retinal degenerative disease, retinitis pigmentosa (RP) is one of the most frequent causes of blindness in adults 20 to 60 years of age [1, 2], leading first to night blindness and mid-peripheral visual field defects during adolescence [1], and then to far peripheral field and central vision abnormalities. A retina with a waxy optic disc, atrophic retinal blood vessels, and bone spicule pigment is typical of RP [3]. Dozens of gene abnormalities on several chromosomes can lead to RP, and these abnormalities can be linked to the Xchromosome and autosomes [1, 2, 4]. Despite the significant scientific progress in terms of genetic research in RP [5, 6], an effective treatment for this serious retinal disease has not yet been found [2], largely because little is known about the encoded protein and metabolic disturbance of the degenerative process [7]. Degeneration of photoreceptors [4], biochemical abnormalities [1, 4, 8], retinal hemodynamic impairment [3] , and micronutrient deficiencies are thought to be involved in the pathogenesis of RP; however, the exact mechanism is not clear. Oxygen is one of the most important metabolites in the retina; it is essential and critical for retinal function, and is associated with a large proportion of retinal blindness [9–11]. However, retinal oxygen conditions in RP remain controversial [11–15]. Some research has suggested that retinal neurons may suffer from oxygen deficiency in RP. Based on this theory, hyperoxic therapy such as hyperbaric oxygen
Graefes Arch Clin Exp Ophthalmol
The study protocol was reviewed and approved by the Medical Ethics Committee of the Zhongshan Ophthalmic Center, Sun Yat-sen University (No. 2013MEKY028), China, and adhered strictly to the principles of the World Medical Association Declaration of Helsinki. After receiving an explanation of the purpose and process of the study, all subjects signed written informed consent before participating in the study.
compromised peripheral and night vision, and reduced ERG b-wave amplitude in severe cases. Inclusion criteria for control subjects were as follows: Chinese (xanthoderm origin), best-corrected visual acuity (BCVA, ETDRS charts)≥75, no ocular diseases, and no history of eye surgery. Exclusion criteria for RP patients and control subjects were as follows: other ocular and/or other pathologies that could affect measurement of retinal oximetry, diabetes, severe cardiovascular or respiratory disease, unstable fixation, oximetry fundus images with poor quality, or refusal to participate in the study. Relative oxygen saturation was measured in retinal blood vessels in both eyes of 68 patients with RP (mean age 41.6± 12.5 years, range 21–68 years) and 136 healthy individuals. There was no significant difference in age or ratio of men to women between RP and control groups, as shown in Table 1. Subjects were assigned to one of two subgroups according to age: Group A (20–40 years) and Group B (> 40 years). All statistics were recorded and compared. Clinical data of the study groups are shown in Table 1. All subjects underwent a complete ophthalmic examination that included BCVA, intraocular pressure (IOP) examination by non-contact tonometry (NCT) (Canon TX-20; Canon Inc., Tokyo, Japan), a slit-lamp examination (Suzhou YZ5S; Suzhou 66 Vision-Tech Co., Ltd., Suzhou, China), and a color fundus image (Topcon TRC-50DX; Topcon Corporation, Tokyo, Japan) . Finger pulse oximetry (Biolight M70; Guangdong Biolight Meditech Co., Ltd., Zhuhai, China), blood pressure, and heart rate (BangPu BF-1100; Shenzhen BangPu Co., Shenzhen, China) were measured prior to retinal oximetry. Scotopic full-field ERGs (UTAS-E 3000 ERG system; LKC Technologies, Inc., Gaithersburg, MD, USA) were examined in 24 RP patients.
Study subjects
Study procedures
Inclusion criteria for RP patients were as follows: Chinese (xanthoderm origin), typical funduscopic appearance for RP,
The Oxymap T1 retinal oximeter (Oxymap ehf., Reykjavik, Iceland) was installed on a fundus camera (Topcon TRC-
(HBO) delivery has been used in some clinical studies to treat RP, and it has achieved encouraging results [15–18]. Conversely, the Boxygen toxicity^ hypothesis [19] holds that hypoxic therapy could alleviate the condition in the later stages of RP [20–22]. As in many other diseases, the premise of appropriate therapy for RP—hypoxic or hyperoxic—is a clear understanding of the normal retinal oxygen distribution and oxygen metabolism abnormalities at various stages of the disease. Thus, measurements of retinal vessel oxygen saturation could aid in assessing disease progression and the effects of treatment. The retinal oximeter is a dual-wavelength retinal camera that can measure retinal vessel oxygen saturation and diameter in vivo [23]. Previous studies have found increased venule saturation and decreased vessel diameter [24] and retinal oxygen saturation to be correlated with structural changes in the eyes of patients with RP [25]. In our study, we aimed to investigate the oxygen saturation and diameter in retinal arterioles and venules in RP patients compared with healthy subjects in different age groups. These differences in age were necessary to understand the pathological changes that occur in RP as the disease progresses.
Methods
Table 1
Clinical data of study groups RP
Age group No. patients Gender, female/male Age, mean±SD, years Intraocular pressure, mean±SD, mmHg Systolic blood pressure, mean±SD, mmHg Diastolic blood pressure, mean±SD, mmHg Pulse, mean±SD Perfusion pressure, mean±SD, mmHg Finger oximetry, mean±SD, %
Group A (20–40) 28 4/3 28.5±4.6 13.2±2.3 119.0±12.5 79.5±12.0 78.4±10.1 57.4±8.4 97.7±1.2
Control Group B (> 40) 40 2/3 50.9±6.6 13.2±2.8 129.2±17.4 85.0±13.0 76.3±12.1 63.1±9.8 97.9±1.2
Group A (20–40) 56 4/3 24.3±2.8 13.6±3.1 108.4±9.2 69.0±6.9 72.0±10.7 50.0±5.4 97.9±0.5
Group B (> 40) 80 2/3 54.5±8.4 13.3±2.2 128.2±16.8 82.2±11.3 75.9±12.6 62.0±9.7 97.3±1.2
Graefes Arch Clin Exp Ophthalmol
50DX; Topcon Corporation, Tokyo, Japan). The oximeter captured images of the retina at two wavelengths simultaneously (570 nm and 600 nm), and then calculated the optical density (OD) of the retinal arterioles and venules at both wavelengths, as shown in Fig 1. Light absorbance can be calculated as follows: OD=log (I0/I), where I0 is the original light intensity and I is the intensity after attenuation due to light absorbance [26–28]. An approximately linear correlation between the optical density ratio (ODR) of the two ODs and hemoglobin oxygen saturation (SatO2) was found in a previous study [27]. The derivation is as follows: SatO2 ¼ ða*ODR þ b þ c*w þ dÞ*100% Calibration parameter: a=−1.28, b=1.24; diameter correction parameters: c=0.0097, d=−0.14; ODR = optical density ratio (dimensionless), w = vessel width. The calibration parameters are those provided by the oximeter manufacturer. Vessel diameter is important for measuring differences in oxygen saturation, and the method of correction was described by Geirsdottir et al. [23, 26, 27]. Subjects’ pupils were dilated with 0.5 % tropicamide (Shenyang Xingji Co., Shenyang, China) for at least 15 minutes before retinal oximetry, until the pupils were dilated to 5.5-6.0 mm in diameter. Retinal images were taken in a darkroom without sunlight and with no illumination other than the light from the camera. The same skilled photographer performed all measurements, using the following standard procedures and parameters: (1) lowest illumination intensity; (2) flash intensity=50 Ws; (3) small aperture and small pupil; (4) consistent angle of gaze, with the optic disc located at the image center; and (5) the same order of photographs (right first, with three images for each eye; the time between images was approximately 1 min, and the clearest image of the retina, with the optic disc in the center, was selected for further analysis). Both eyes in each RP patient and control subject were studied. The pseudocolor fundus image can provide a quick overview of oxygen saturation of retinal vessels. Measurements were made from the inner circle (close to the edge of the optic disc) to the outer circle (400 pixels from the inner circle), as shown in Fig 1. All arterioles and venules in this area 6 pixels or greater in diameter were measured. In this study, all arteriolar or venular segments were selected at the same time, and the mean oxygen saturation values and vascular width of all segments of arterioles or venules were automatically analyzed using data analysis software (Oxymap Analyzer version 2.4; Oxymap ehf.). This analysis was performed based on the ODR, and the means and standard deviations of the fundus oxygen saturation measurements were determined.
Ocular perfusion pressure (OPP), which is the driving force in ocular blood circulation, is affected by systemic blood pressure and IOP. The equation for OPP is: OPP ¼ 2=3ð2=3 BPdiast þ 1=3 BPsystÞIOP BPdiast and BPsyst are the diastolic and systolic blood pressure values, respectively [23, 29]. The disease duration of RP was defined from the onset of symptoms. Disease duration and ERG b-wave amplitude were each divided into four degrees, as shown in Tables 2 and 3. Statistical analysis In this study, one eye of each subject was randomly selected for statistical analysis. The Student's t test was used to analyze the mean saturation and vessel diameter of selected retinal arterioles and venules and arteriovenous differences between RP and normal subjects by age. The Spearman test was used to analyze the correlation of oxygen saturation with visual acuity (VA), disease duration, and ERG b-wave amplitude. Data are expressed as means±standard deviations (SD), and a p value 40 years old). Oxygen saturation was measured along the main retinal vessels within 400 pixels of the edge of the optic disc. The colors in the oximetry images indicate relative oxygen saturation (%) in the retinal vessels. Note that oxygen saturation of the retinal arterioles was increased in Group A, while in Group B it was decreased
Graefes Arch Clin Exp Ophthalmol Table 2 Degrees of disease duration of RP
Degree
Disease duration of RP (years)
Degree 0 Degree 1 Degree 2 Degree 3
< 10 10–20 20–30 ≥ 30
The mean diameter of retinal arterioles (AD) and venules (VD) was significantly smaller in RP patients of both age groups than in healthy controls (p=0.000, p=0.000). No significant differences were found in AD or VD between the two age groups (p=0.757, p=0.340). For patients with RP, the mean AS and AVS were negatively correlated with disease duration (r=−0.323, p=0.007; r= −0.276, p=0.023). Furthermore, the mean AS had a nonsignificant positive correlation with the b-wave (r=0.356, p=0.088). The mean AD and VD were significantly negatively correlated with disease duration (r=−0. 329, p=0.006; r= −0. 257, p=0.034, respectively) and positively correlated with the b-wave (r=0. 502, p=0.012; r=0. 428, p=0.037, respectively). There was also a statistically significant positive correlation between AD and VA (r=0.257, p=0.034), as shown in Table 5. Figure 3 shows that at the early stage of RP (< 10 years), the mean AS did not decrease substantially, while the variation in AS increased. At a later stage (≥ 10 years), and with a decreased b-wave, the mean AS decreased but the variation in AS still increased.
Discussion Mean AS and AVS were significantly higher in patients with RP than in control subjects in Group A, and significantly lower than in controls in Group B. No significant differences were found in mean VS between RP patients and healthy controls in either age group. These results are different from those in previous studies, one of which examined 21 eyes of 11 RP patients and found that AS and VS were higher and AVS was lower in RP patients compared to control subjects [25]. Another study of 10 patients with advanced RP found
Table 3 Degrees of ERG b-wave with RP
Degree
ERG b-wave
Degree 0 Degree 1 Degree 2 Degree 3
No wave Severe reduction Moderate reduction Minor reduction
increased VS in RP, but no significant difference in AS between RP and control subjects [24]. However, no previous studies have investigated retinal vessel oxygen saturation and diameter in different age groups. Differences between our study and previous research may be due to the age grouping, the severity of the disease, and the limited sample sizes in previous studies. Increased AS and AVS levels were found in RP patients in Group A, and no significant difference was found in VS. Several possible factors might have contributed to these changes. First, the decreased vessel diameters in RP may lead to high measured oxygen saturation values, as reported in a previous study [25]. Because the retinal oximeter was designed based on healthy persons, the correction for the Oxymap software may be not perfect [23, 26]. Second, the higher AS may be caused by reduced oxygen diffusion from the retinal circulation in RP. Higher AS has also been observed in diabetic retinopathy, and a possible mechanism was the shunting of blood within the retinal circulation [30]. We are not aware of any reports of blood shunting in RP, but a histopathologic study found a marked thickening of the basement membrane around the retinal venules and capillaries in RP [31], which may result in a decrease in oxygen diffusion from the capillaries to the tissue and an increase in oxygen saturation of retinal vessels. In addition, as the tissue immediately adjacent to the retinal arterioles is normally supplied by diffusion from the arterioles due to the capillaryfree zone around the arterioles, if this tissue consumes less oxygen because of atrophy, the saturation in the arterioles could rise slightly. Finally, the increased AVS may be physiological compensation for relatively insufficient oxygen delivery in the retina in the early stage of RP. Despite the loss of cells that has been found in all layers of the retina, there are still plenty of histologically intact inner retinal neurons in the early stage of RP [32]. Previous studies [33–36] showed that retinal blood flow was significantly reduced in RP patients compared to healthy subjects, even in the early stages. These surviving neurons may suffer from hypoxia and may take more oxygen per unit volume of blood, resulting in higher AVS. Hyperoxic therapy has been used to treat RP in several clinical studies, and significant improvements have been reported with VA, visual field, and ERG responses, as well as a slowing of the retinal degenerative process [15–18]. In animal models of retinal degeneration, hyperoxic therapy has been shown to reduce photoreceptor and neuron death in the early stage of RP, called the Bcritical period^ [37–39]. Decreased AS levels were found in RP patients in Group B, and no significant difference was found in VS, thus resulting in a lower AVS. The oxygen delivery from the retinal circulation to the retinal tissue can be estimated from AVS multiplied by blood flow, which was expected to be reduced in RP patients in Group B. One reason for the decrease in oxygen
Graefes Arch Clin Exp Ophthalmol Table 4
Oxygen saturation and vessel diameter in RP patients and control subjects in different age groups All RP
Oxygen saturation (%) Arterioles (AS) 94.5±13.6 Venules (VS) 61.8±7.5 AV difference (AVS) 32.8±15.4 Vessel diameter (pixels) Arterioles (AD) 10.4±1.7 Venules (VD) 13.8±1.9
Group A (20–40 years)
Group B (> 40 years)
RP
Control
P (t test)
RP
Control
P (t test)
RP
Control
P (t test)
P’(t test)
95.4±4.7 61.3±5.0 34.1±5.3
0.609 0.665 0.498
105. 5±9.4 62.0±4. 7 43.5±9.6
94.5±4.4 62.0±5.0 32.7±4.9
0.000* 0.875 0.000*
86.8±10.3 61.6±9.1 25.2±14.3
96.0±4.8 61.0±5.1 35.1±5.3
0.000* 0.689 0.000*
0.000* 0.824 0.000*
13.3±1.2 15.8±1.4
0.000* 0.000*
10.5±1.7 13.5±1.4
13.5±1.2 16.0±1.2
0.000* 0.000*
10.4±1.8 14.0±2.2
13.2±1.3 15.7±1.4
0.000* 0.000*
0.757 0.340
Oxygen saturation and vessel diameter (mean±SD) are shown for all groups, with p values for comparison of RP patients to healthy individuals. No significant difference was found in mean AS, VS or AVS between all RP patients and all healthy controls. An increase in AS and AVS in Group A and a decrease in Group B can be seen in the eyes of RP patients. No significant difference was found in mean VS between RP patients and healthy controls in either age group. For both age groups, mean AD and VD were significantly smaller in patients with RP than in normal individuals The P' values for comparison of oxygen saturation and vessel diameter in RP patients between the two age groups show that patients in Group B had significantly lower AS and AVS than those in Group A, while no significant difference was found in AD or VD between the two age groups *Significant difference in oxygen saturation and vessel diameter between RP patients and healthy subjects by age group according to the Student's t test (p
RETINAL DISORDERS
Retinal vessel oxygen saturation and vessel diameter in retinitis pigmentosa at various ages Yao Zong 1 & Leilei Lin 1 & Changxian Yi 1 & Xia Huang 1 & Yue Fu 1 & Yanmin Dong 1 & Xiaobing Qian 1 & Yujie Li 1 & Qianying Gao 1
Received: 11 December 2014 / Revised: 16 April 2015 / Accepted: 28 April 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Purpose This study was conducted to determine whether oxygen saturation and retinal blood vessel diameter are affected by retinitis pigmentosa (RP) at various ages. Methods Relative oxygen saturation was measured in retinal blood vessels in 68 RP patients and 136 normal subjects using the Oxymap T1 retinal oximeter. Subjects were divided into two age groups: Group A (20–40 years) and Group B (> 40 years). One randomly selected eye of each subject was used for statistical analysis. Student's t tests were used to analyze the mean saturation and diameter of retinal arterioles and venules and arteriovenous differences between RP and normal subjects in the two age groups. A Spearman test was used to analyze the correlation of mean saturation of retinal arterioles (AS) and arteriovenous differences (AVS) with visual acuity, disease duration, and electroretinogram (ERG) b-wave amplitude in patients with RP. Results AS was significantly higher in patients with RP (105.5±9.4 %) than in normal subjects (94.5±4.4 %, p= 0.000) in Group A, while in Group B, AS was significantly lower in RP patients (86.8±10.3 %) than in healthy subjects (96.0±4.8 %, p=0.000). Vessel diameter was smaller in RP patients than in normal subjects. AS and AVS showed a negative correlation with disease duration and a tendency toward positive correlation with ERG b-wave in patients with RP. Conclusions The shifting characteristics of retinal vessel oxygen saturation suggest that the pathological mechanism of
* Qianying Gao [email protected] 1
State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University,, 54 Xianlie Road, Guangzhou, Guangdong 510060, China
retinal oxygen metabolic disorder differs by age in patients with RP. Keywords Hypoxia . Oximetry . Oxygen saturation . Retina . Retinitis pigmentosa
Introduction As an inherited retinal degenerative disease, retinitis pigmentosa (RP) is one of the most frequent causes of blindness in adults 20 to 60 years of age [1, 2], leading first to night blindness and mid-peripheral visual field defects during adolescence [1], and then to far peripheral field and central vision abnormalities. A retina with a waxy optic disc, atrophic retinal blood vessels, and bone spicule pigment is typical of RP [3]. Dozens of gene abnormalities on several chromosomes can lead to RP, and these abnormalities can be linked to the Xchromosome and autosomes [1, 2, 4]. Despite the significant scientific progress in terms of genetic research in RP [5, 6], an effective treatment for this serious retinal disease has not yet been found [2], largely because little is known about the encoded protein and metabolic disturbance of the degenerative process [7]. Degeneration of photoreceptors [4], biochemical abnormalities [1, 4, 8], retinal hemodynamic impairment [3] , and micronutrient deficiencies are thought to be involved in the pathogenesis of RP; however, the exact mechanism is not clear. Oxygen is one of the most important metabolites in the retina; it is essential and critical for retinal function, and is associated with a large proportion of retinal blindness [9–11]. However, retinal oxygen conditions in RP remain controversial [11–15]. Some research has suggested that retinal neurons may suffer from oxygen deficiency in RP. Based on this theory, hyperoxic therapy such as hyperbaric oxygen
Graefes Arch Clin Exp Ophthalmol
The study protocol was reviewed and approved by the Medical Ethics Committee of the Zhongshan Ophthalmic Center, Sun Yat-sen University (No. 2013MEKY028), China, and adhered strictly to the principles of the World Medical Association Declaration of Helsinki. After receiving an explanation of the purpose and process of the study, all subjects signed written informed consent before participating in the study.
compromised peripheral and night vision, and reduced ERG b-wave amplitude in severe cases. Inclusion criteria for control subjects were as follows: Chinese (xanthoderm origin), best-corrected visual acuity (BCVA, ETDRS charts)≥75, no ocular diseases, and no history of eye surgery. Exclusion criteria for RP patients and control subjects were as follows: other ocular and/or other pathologies that could affect measurement of retinal oximetry, diabetes, severe cardiovascular or respiratory disease, unstable fixation, oximetry fundus images with poor quality, or refusal to participate in the study. Relative oxygen saturation was measured in retinal blood vessels in both eyes of 68 patients with RP (mean age 41.6± 12.5 years, range 21–68 years) and 136 healthy individuals. There was no significant difference in age or ratio of men to women between RP and control groups, as shown in Table 1. Subjects were assigned to one of two subgroups according to age: Group A (20–40 years) and Group B (> 40 years). All statistics were recorded and compared. Clinical data of the study groups are shown in Table 1. All subjects underwent a complete ophthalmic examination that included BCVA, intraocular pressure (IOP) examination by non-contact tonometry (NCT) (Canon TX-20; Canon Inc., Tokyo, Japan), a slit-lamp examination (Suzhou YZ5S; Suzhou 66 Vision-Tech Co., Ltd., Suzhou, China), and a color fundus image (Topcon TRC-50DX; Topcon Corporation, Tokyo, Japan) . Finger pulse oximetry (Biolight M70; Guangdong Biolight Meditech Co., Ltd., Zhuhai, China), blood pressure, and heart rate (BangPu BF-1100; Shenzhen BangPu Co., Shenzhen, China) were measured prior to retinal oximetry. Scotopic full-field ERGs (UTAS-E 3000 ERG system; LKC Technologies, Inc., Gaithersburg, MD, USA) were examined in 24 RP patients.
Study subjects
Study procedures
Inclusion criteria for RP patients were as follows: Chinese (xanthoderm origin), typical funduscopic appearance for RP,
The Oxymap T1 retinal oximeter (Oxymap ehf., Reykjavik, Iceland) was installed on a fundus camera (Topcon TRC-
(HBO) delivery has been used in some clinical studies to treat RP, and it has achieved encouraging results [15–18]. Conversely, the Boxygen toxicity^ hypothesis [19] holds that hypoxic therapy could alleviate the condition in the later stages of RP [20–22]. As in many other diseases, the premise of appropriate therapy for RP—hypoxic or hyperoxic—is a clear understanding of the normal retinal oxygen distribution and oxygen metabolism abnormalities at various stages of the disease. Thus, measurements of retinal vessel oxygen saturation could aid in assessing disease progression and the effects of treatment. The retinal oximeter is a dual-wavelength retinal camera that can measure retinal vessel oxygen saturation and diameter in vivo [23]. Previous studies have found increased venule saturation and decreased vessel diameter [24] and retinal oxygen saturation to be correlated with structural changes in the eyes of patients with RP [25]. In our study, we aimed to investigate the oxygen saturation and diameter in retinal arterioles and venules in RP patients compared with healthy subjects in different age groups. These differences in age were necessary to understand the pathological changes that occur in RP as the disease progresses.
Methods
Table 1
Clinical data of study groups RP
Age group No. patients Gender, female/male Age, mean±SD, years Intraocular pressure, mean±SD, mmHg Systolic blood pressure, mean±SD, mmHg Diastolic blood pressure, mean±SD, mmHg Pulse, mean±SD Perfusion pressure, mean±SD, mmHg Finger oximetry, mean±SD, %
Group A (20–40) 28 4/3 28.5±4.6 13.2±2.3 119.0±12.5 79.5±12.0 78.4±10.1 57.4±8.4 97.7±1.2
Control Group B (> 40) 40 2/3 50.9±6.6 13.2±2.8 129.2±17.4 85.0±13.0 76.3±12.1 63.1±9.8 97.9±1.2
Group A (20–40) 56 4/3 24.3±2.8 13.6±3.1 108.4±9.2 69.0±6.9 72.0±10.7 50.0±5.4 97.9±0.5
Group B (> 40) 80 2/3 54.5±8.4 13.3±2.2 128.2±16.8 82.2±11.3 75.9±12.6 62.0±9.7 97.3±1.2
Graefes Arch Clin Exp Ophthalmol
50DX; Topcon Corporation, Tokyo, Japan). The oximeter captured images of the retina at two wavelengths simultaneously (570 nm and 600 nm), and then calculated the optical density (OD) of the retinal arterioles and venules at both wavelengths, as shown in Fig 1. Light absorbance can be calculated as follows: OD=log (I0/I), where I0 is the original light intensity and I is the intensity after attenuation due to light absorbance [26–28]. An approximately linear correlation between the optical density ratio (ODR) of the two ODs and hemoglobin oxygen saturation (SatO2) was found in a previous study [27]. The derivation is as follows: SatO2 ¼ ða*ODR þ b þ c*w þ dÞ*100% Calibration parameter: a=−1.28, b=1.24; diameter correction parameters: c=0.0097, d=−0.14; ODR = optical density ratio (dimensionless), w = vessel width. The calibration parameters are those provided by the oximeter manufacturer. Vessel diameter is important for measuring differences in oxygen saturation, and the method of correction was described by Geirsdottir et al. [23, 26, 27]. Subjects’ pupils were dilated with 0.5 % tropicamide (Shenyang Xingji Co., Shenyang, China) for at least 15 minutes before retinal oximetry, until the pupils were dilated to 5.5-6.0 mm in diameter. Retinal images were taken in a darkroom without sunlight and with no illumination other than the light from the camera. The same skilled photographer performed all measurements, using the following standard procedures and parameters: (1) lowest illumination intensity; (2) flash intensity=50 Ws; (3) small aperture and small pupil; (4) consistent angle of gaze, with the optic disc located at the image center; and (5) the same order of photographs (right first, with three images for each eye; the time between images was approximately 1 min, and the clearest image of the retina, with the optic disc in the center, was selected for further analysis). Both eyes in each RP patient and control subject were studied. The pseudocolor fundus image can provide a quick overview of oxygen saturation of retinal vessels. Measurements were made from the inner circle (close to the edge of the optic disc) to the outer circle (400 pixels from the inner circle), as shown in Fig 1. All arterioles and venules in this area 6 pixels or greater in diameter were measured. In this study, all arteriolar or venular segments were selected at the same time, and the mean oxygen saturation values and vascular width of all segments of arterioles or venules were automatically analyzed using data analysis software (Oxymap Analyzer version 2.4; Oxymap ehf.). This analysis was performed based on the ODR, and the means and standard deviations of the fundus oxygen saturation measurements were determined.
Ocular perfusion pressure (OPP), which is the driving force in ocular blood circulation, is affected by systemic blood pressure and IOP. The equation for OPP is: OPP ¼ 2=3ð2=3 BPdiast þ 1=3 BPsystÞIOP BPdiast and BPsyst are the diastolic and systolic blood pressure values, respectively [23, 29]. The disease duration of RP was defined from the onset of symptoms. Disease duration and ERG b-wave amplitude were each divided into four degrees, as shown in Tables 2 and 3. Statistical analysis In this study, one eye of each subject was randomly selected for statistical analysis. The Student's t test was used to analyze the mean saturation and vessel diameter of selected retinal arterioles and venules and arteriovenous differences between RP and normal subjects by age. The Spearman test was used to analyze the correlation of oxygen saturation with visual acuity (VA), disease duration, and ERG b-wave amplitude. Data are expressed as means±standard deviations (SD), and a p value 40 years old). Oxygen saturation was measured along the main retinal vessels within 400 pixels of the edge of the optic disc. The colors in the oximetry images indicate relative oxygen saturation (%) in the retinal vessels. Note that oxygen saturation of the retinal arterioles was increased in Group A, while in Group B it was decreased
Graefes Arch Clin Exp Ophthalmol Table 2 Degrees of disease duration of RP
Degree
Disease duration of RP (years)
Degree 0 Degree 1 Degree 2 Degree 3
< 10 10–20 20–30 ≥ 30
The mean diameter of retinal arterioles (AD) and venules (VD) was significantly smaller in RP patients of both age groups than in healthy controls (p=0.000, p=0.000). No significant differences were found in AD or VD between the two age groups (p=0.757, p=0.340). For patients with RP, the mean AS and AVS were negatively correlated with disease duration (r=−0.323, p=0.007; r= −0.276, p=0.023). Furthermore, the mean AS had a nonsignificant positive correlation with the b-wave (r=0.356, p=0.088). The mean AD and VD were significantly negatively correlated with disease duration (r=−0. 329, p=0.006; r= −0. 257, p=0.034, respectively) and positively correlated with the b-wave (r=0. 502, p=0.012; r=0. 428, p=0.037, respectively). There was also a statistically significant positive correlation between AD and VA (r=0.257, p=0.034), as shown in Table 5. Figure 3 shows that at the early stage of RP (< 10 years), the mean AS did not decrease substantially, while the variation in AS increased. At a later stage (≥ 10 years), and with a decreased b-wave, the mean AS decreased but the variation in AS still increased.
Discussion Mean AS and AVS were significantly higher in patients with RP than in control subjects in Group A, and significantly lower than in controls in Group B. No significant differences were found in mean VS between RP patients and healthy controls in either age group. These results are different from those in previous studies, one of which examined 21 eyes of 11 RP patients and found that AS and VS were higher and AVS was lower in RP patients compared to control subjects [25]. Another study of 10 patients with advanced RP found
Table 3 Degrees of ERG b-wave with RP
Degree
ERG b-wave
Degree 0 Degree 1 Degree 2 Degree 3
No wave Severe reduction Moderate reduction Minor reduction
increased VS in RP, but no significant difference in AS between RP and control subjects [24]. However, no previous studies have investigated retinal vessel oxygen saturation and diameter in different age groups. Differences between our study and previous research may be due to the age grouping, the severity of the disease, and the limited sample sizes in previous studies. Increased AS and AVS levels were found in RP patients in Group A, and no significant difference was found in VS. Several possible factors might have contributed to these changes. First, the decreased vessel diameters in RP may lead to high measured oxygen saturation values, as reported in a previous study [25]. Because the retinal oximeter was designed based on healthy persons, the correction for the Oxymap software may be not perfect [23, 26]. Second, the higher AS may be caused by reduced oxygen diffusion from the retinal circulation in RP. Higher AS has also been observed in diabetic retinopathy, and a possible mechanism was the shunting of blood within the retinal circulation [30]. We are not aware of any reports of blood shunting in RP, but a histopathologic study found a marked thickening of the basement membrane around the retinal venules and capillaries in RP [31], which may result in a decrease in oxygen diffusion from the capillaries to the tissue and an increase in oxygen saturation of retinal vessels. In addition, as the tissue immediately adjacent to the retinal arterioles is normally supplied by diffusion from the arterioles due to the capillaryfree zone around the arterioles, if this tissue consumes less oxygen because of atrophy, the saturation in the arterioles could rise slightly. Finally, the increased AVS may be physiological compensation for relatively insufficient oxygen delivery in the retina in the early stage of RP. Despite the loss of cells that has been found in all layers of the retina, there are still plenty of histologically intact inner retinal neurons in the early stage of RP [32]. Previous studies [33–36] showed that retinal blood flow was significantly reduced in RP patients compared to healthy subjects, even in the early stages. These surviving neurons may suffer from hypoxia and may take more oxygen per unit volume of blood, resulting in higher AVS. Hyperoxic therapy has been used to treat RP in several clinical studies, and significant improvements have been reported with VA, visual field, and ERG responses, as well as a slowing of the retinal degenerative process [15–18]. In animal models of retinal degeneration, hyperoxic therapy has been shown to reduce photoreceptor and neuron death in the early stage of RP, called the Bcritical period^ [37–39]. Decreased AS levels were found in RP patients in Group B, and no significant difference was found in VS, thus resulting in a lower AVS. The oxygen delivery from the retinal circulation to the retinal tissue can be estimated from AVS multiplied by blood flow, which was expected to be reduced in RP patients in Group B. One reason for the decrease in oxygen
Graefes Arch Clin Exp Ophthalmol Table 4
Oxygen saturation and vessel diameter in RP patients and control subjects in different age groups All RP
Oxygen saturation (%) Arterioles (AS) 94.5±13.6 Venules (VS) 61.8±7.5 AV difference (AVS) 32.8±15.4 Vessel diameter (pixels) Arterioles (AD) 10.4±1.7 Venules (VD) 13.8±1.9
Group A (20–40 years)
Group B (> 40 years)
RP
Control
P (t test)
RP
Control
P (t test)
RP
Control
P (t test)
P’(t test)
95.4±4.7 61.3±5.0 34.1±5.3
0.609 0.665 0.498
105. 5±9.4 62.0±4. 7 43.5±9.6
94.5±4.4 62.0±5.0 32.7±4.9
0.000* 0.875 0.000*
86.8±10.3 61.6±9.1 25.2±14.3
96.0±4.8 61.0±5.1 35.1±5.3
0.000* 0.689 0.000*
0.000* 0.824 0.000*
13.3±1.2 15.8±1.4
0.000* 0.000*
10.5±1.7 13.5±1.4
13.5±1.2 16.0±1.2
0.000* 0.000*
10.4±1.8 14.0±2.2
13.2±1.3 15.7±1.4
0.000* 0.000*
0.757 0.340
Oxygen saturation and vessel diameter (mean±SD) are shown for all groups, with p values for comparison of RP patients to healthy individuals. No significant difference was found in mean AS, VS or AVS between all RP patients and all healthy controls. An increase in AS and AVS in Group A and a decrease in Group B can be seen in the eyes of RP patients. No significant difference was found in mean VS between RP patients and healthy controls in either age group. For both age groups, mean AD and VD were significantly smaller in patients with RP than in normal individuals The P' values for comparison of oxygen saturation and vessel diameter in RP patients between the two age groups show that patients in Group B had significantly lower AS and AVS than those in Group A, while no significant difference was found in AD or VD between the two age groups *Significant difference in oxygen saturation and vessel diameter between RP patients and healthy subjects by age group according to the Student's t test (p
