Evaluation of visual field parameters in patients with chronic obstructive pulmonary disease

Authors


Helin Deniz Demir, MD
Gaziosmanpasa Universitesi Tıp Fakültesi
Göz Hastalıkları ABD
60100 Tokat
Turkey
Tel: + 90 532 776 15 21
Fax: + 90 356 213 31 79
Email: helindeniz@hotmail.com

Abstract.

Purpose:  To evaluate the effects of chronic obstructive pulmonary disease (COPD) on retina and optic nerve.

Methods:  Thirty-eight patients with COPD and 29 healthy controls, totally 67 subjects, were included in the study. Visual evoked potentials (VEP) and visual field assessment (both standard achromatic perimetry (SAP) and short-wavelength automated perimetry (SWAP)) were performed on each subject after ophthalmological, neurological and pulmonary examinations.

Results:  Mean deviation (MD), pattern standard deviation (PSD) and corrected pattern standard deviation (CPSD) were significantly different between patient and control groups as for both SAP and SWAP measurements (p = 0.001, 0.019, 0.009 and p = 0.004,0.019, 0.031, respectively). Short-term fluctuation (SF) was not statistically different between the study and the control groups (p = 0.874 and 0.694, respectively). VEP P100 latencies were significantly different between patients with COPD and the controls (p = 0.019).

Conclusion:  Chronic obstructive pulmonary disease is a systemic disease, and hypoxia in COPD seems to affect the retina and the optic nerve.

Introduction

Chronic obstructive pulmonary disease (COPD) is characterized by abnormal inflammatory response of the lungs to noxious particles and gases resulting in progressive airflow limitation (Global Initiative for Chronic Obstructive Lung Disease 2006). Today, it is accepted as a systemic disease that affects all bodily systems (Agusti & Soriano 2008). Tissue hypoxia, smoking and inflammatory cytokines are important in the development of the systemic effects of COPD (Vernooy et al. 2002; Gan et al. 2004).

Nervous system is also affected during the disease process. Appenzeller et al. (1968) firstly reported Wallerian degeneration and demyelinization of the peripheral nerves in patients with COPD (Pfeiffer et al. 1990). Afterwards, the role of polyneuropathy (PNP) in COPD has been shown in various studies (Ozge & Atiş 2001; Agrawal et al. 2007). Hypoxaemia has been proposed as the major factor for the development of polyneuropathy (Pfeiffer et al. 1990; Ozge & Atiş 2001; Ozge et al. 2005). Hypoxaemia may cause neuropathy by its direct effect on the nerve fibres or by increasing the impact of neurotoxic factors. Besides hypoxaemia, smoking, hypercapnia, advanced age, and malnutrition can be contributing factors in COPD-related neuropathy (Pfeiffer et al. 1990; Ozge & Atiş 2001; Ozge et al. 2005). Studies indicating axonal and demyelinating changes have been reported in patients with COPD as demonstrated by electrophysiological methods (Ozge et al. 2005). Only a few studies have been carried out on the role of the cranial neuropathies in COPD. Ozge et al. studied visual evoked potentials and reported visual evoked potentials (VEP) latencies and amplitude abnormalities in these patients. To the best of our knowledge, the relation between COPD and visual field has not been studied. The aim of this study is to evaluate the optic nerve and retinal involvement and its effect on the visual field in patients with COPD.

Methods

Clinically stable 38 patients with COPD and 29 healthy controls, totally 67 subjects, were included in this prospective, case–control study. The diagnoses of the patients with COPD were established according to the GOLD Guidelines (Global Initiative for Chronic Obstructive Lung Disease 2006). Patients with asthma, bronchiectasis, tuberculosis, interstitial lung diseases, chronic diseases like renal, heart and hepatic failure, uncontrolled diabetes and hypertension, hypo- or hyperthyroidism, malnutrition, neurological disorders (stroke, dementias, multiple sclerosis, epilepsy or sleep disorders), intracranial or intraorbital space occupying lesions, colour vision defects, ocular trauma, optic disc and retinal pathologies, cataract, miosis, glaucoma, high refractive errors were excluded from the study.

After obtaining a detailed medical history, all symptoms and their duration, and smoking habits were recorded. Arterial blood gas analyses, posteroanterior and left lateral chest graphs were performed only in patients with COPD while pulmonary function test (PFTs) in the study group. The control group was chosen from subjects with normal PFT test results and without any evidence of previous or current obstructive pulmonary disease and a history of smoking.

After a full ophthalmological examination, visual field analysis and visual evoked potentials (VEPs) were elicited from all subjects. All subjects had normal visual acuity results as detected using Snellen charts.

During VEP recordings, all patients were seated comfortably in a semi-darkened room, and exposed to the stimuli coming from a television monitor 1 m away from the tested eye. Vision was central and monocular. All subjects were asked to keep their eye on a central fixation point. Recordings were performed using Medelec Synergy EMG machine with an analysis time of 500 ms and a sweep speed of 50 ms/s. The mean luminance was 43.3 cd/m2, and the contrast was 100%. Low- and high-frequency filter settings were predetermined at 1.0 and 100 Hz, respectively. The international 10–20 system was used to insert electrodes. VEPs were recorded from five channels (T5-Cz, O1-Cz, Oz-Cz, O2-Cz and T6-Cz). The ground electrode was attached on the forearm. VEPs were obtained by monocular checkerboard pattern reversal stimulation. An average of 200 runs was taken, and each run was checked for reproducibility by a second waveform stored in the memory system. For each eye, the first prominent positive (downward deflection) peak P100 was obtained. The latencies to the peak P100 and peak-to-peak amplitude of P100 were measured.

Both eyes of each subject were evaluated by a central 30-2 full threshold program of the Humprey Visual Field analyser II model 750i (Humphrey Instruments, San Leandro, CA, USA) with standard achromatic perimetry (SAP) (using Goldman size III stimulus) and short-wavelength automated perimetry (SWAP) (using Goldmann size V stimulus). Each test was performed by the same technician by giving a relaxing period between examinations of each eye. Only one eye of each subject with reliable visual field parameters (fixation loss, <20%; false positive rate, <33%; and false negative rate, <33%) was included in the study. Both visual field analyses were repeated in patients with suspected field defects. Global indices [mean deviation (MD), pattern standard deviation (PSD), short-term fluctuation (SF) and corrected pattern standard deviation (CPSD)] were evaluated. The study was carried out in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. Written informed consent was obtained from all participants.

T-test for independent samples was used for comparing variables between groups. Whenever the data were not normally distributed as evaluated by Levene homogeneity test of variance, Mann–Whitney U-test was used. Simple correlation analysis was performed by Pearson correlation coefficient method.

Results

Thirty-eight subjects with COPD and 29 healthy controls, totally 67 patients, were included in the study. Demographic characteristics and PFTs of cases were summarized in Table 1. The best corrected visual acuity was 20/20 in patients and controls. The mean refractive error of right and left eyes of patients with COPD and healthy controls was not significantly different between groups. (0.78 ± 0.35 D, 0.78 ± 0.35 D, 0.71 ± 0.35 D, 0.72 ± 0.35 D, respectively, p = 0.397) The mean intraocular pressure (IOP) and cup/disc (c/d) ratio were shown in Table 2. Three of patients with COPD (7.8%) were non-smokers, 24 of them (63.1%) quitted smoking within the last year, and 11 of them (28.9%) were current smokers. The control group consisted of 29 healthy non-smokers.

Table 1.   Demographic and clinical data of the study groups.
ParametersPatients with COPD (= 38)Controls (= 29)tzp
  1. FEV1 = Forced expiratory volume in 1-second; FEV1(%) = percentage of predicted values FEV1; FVC = forced expiratory vital capacity; MMEF = maximum mid-expiratory flow; pH = power of hydrogen; PaO2 = partial pressure of oxygen in the blood; PaCO2 = partial pressure of carbon dioxide in the blood; COPD = chronic obstructive pulmonary disease.

  2. t:Statistical value (for comparing independent samples).

  3. z:Statistical value (whenever the data were not normally distrubuted, z-value was used instead of t) p ≤ 0.05.

Age (years)59.55 ± 8.6358.79 ± 5.470.414 0.680
Disease duration (years)5.73 ± 6.73   
Smoking status (pack/year)37.92 ± 20.98   
FEV1 (lt)1.66 ± 0.572.97 ± 0.618.941 0.0001
FEV1 (% of predicted)57.11 ± 16.70101.64 ± 17.7710.518 0.0001
FVC (lt)2.79 ± 0.673.69 ± 0.805.012 0.0001
FVC (% of predicted)76.15 ± 15.42103.13 ± 18.266.550 0.0001
FEV1/FVC ratio58.24 ± 8.9780.80 ± 4.45 −6.9740.0001
MMEF (lt)0.82 ± 0.432.77 ± 0.85 −6.8340.0001
MMEF (% of predicted)25.41 ± 11.4779.99 ± 24.17 −6.8340.0001
pH7.40 ± 0.02   
PaO2 (mmHg)67.28 ± 7.2   
PaCO2 (mmHg)38.58 ± 5.3   
Table 2.   The comparison between IOP and cup/disc (c/d) ratio of the study groups.
 Patients with COPD (n = 38)Controls (= 29)tp
  1. RE = right eye; LE = left eye; c/d ratio = cup/disc ratio; IOP = intraocular pressure; COPD = chronic obstructive pulmonary disease.

  2. t:Statistical value (for comparing independent samples), p ≤ 0.05.

c/d ratio (RE)0.18 ± 0.040.18 ± 0.0040.0160.988
c/d ratio (LE)0.18 ± 0.040.18 ± 0.0.040.2370.813
IOP (RE) mmHg13.0 ± 1.2312.89 ± 1.340.3270.744
IOP (LE) mmHg13.0 ± 1.2012.89 ± 1.340.3310.742

Mean deviation, PSD and CPSD were significantly different between patients and controls based on both SAP and SWAP measurements (p = 0.001, 0.019, 0.009 and p = 0.004, 0.019, 0.031, respectively). SF was not statistically different between the study groups (p = 0.874 and 0.694, respectively) (Table 3). There was no significant correlation between exposure to tobacco and visual field parameters in patients with COPD (Table 4). VEP P100 latency was significantly different between patients with COPD and the controls (p = 0.019) (Table 5).

Table 3.   The comparison between the visual field parameters in patients with chronic obstructive pulmonary disease (COPD) and controls.
ParametersPatients with COPD (= 38)Controls (= 29)tp
  1. SAP = Standard achromatic perimetry; SWAP = short-wavelength automated perimetry; MD = mean deviation; PSD = pattern standard deviation; SF = short-term fluctuation; CPSD = corrected pattern standard deviation.

  2. t: Statistical value (for comparing independent samples), p ≤ 0.05.

SAP
 MD−5.90 ± 3.30−3.42 ± 2.493.3730.001
 PSD4.07 ± 1.913.07 ± 1.302.4120.019
 SF2.10 ± 1.012.06 ± 1.030.1590.874
 CPSD3.04 ± 2.161.80 ± 1.382.6840.009
SWAP
 MD−9.99 ± 4.09−6.74 ± 4.932.9470.004
 PSD4.27 ± 1.173.59 ± 1.122.3960.019
 SF2.55 ± 0.882.46 ± 1.050.3950.694
 CPSD2.74 ± 1.731.84 ± 1.532.1980.031
Table 4.   The correlation between cigarette smoking and visual field parameters.
Parametersrp
  1. SAP = standard achromatic perimetry; SWAP = short-wavelength automated perimetry; MD = mean deviation; PSD = pattern standard deviation; SF = short-term fluctuation; CPSD = corrected pattern standard deviation.

  2. r: correlation coefficient, p ≤ 0.05.

SAP
 MD−0.1600.338
 PSD0.0280.867
 SF0.2560.121
 CPSD−0.0680.685
SWAP
 MD0.0050.974
 PSD0.0150.931
 SF0.1800.281
 CPSD−0.0630.709
Table 5.   The comparison of the visual evoked potential parameters of the study groups.
ParametersPatients with COPD (= 38)Controls (= 29)tp
  1. COPD = chronic obstructive pulmonary disease.

  2. t: Statistical value (for comparing independent samples), p ≤ 0.05.

P100 latency (ms)112.88 ± 11.54107.18 ± 5.412.4110.019
P100 amplitude (μV)7.04 ± 3.146.97 ± 3.440.0800.937

Discussion

Chronic obstructive pulmonary disease is a disease that affects not only the lungs but also whole body (Agusti & Soriano 2008). Polyneuropathy has already been studied in various studies, and its incidence was clinically reported to be 7–88%. Electrophysiological studies evaluating evoked potentials have shown latency anomalies in patients with COPD (Kayacan et al.2001). Ozge et al. (2005) found optic nerve involvement, possibly as part of a polyneuropathy, to be common in COPD, as well. We also found optic nerve VEP P100 latency to be prolonged in patients with COPD. Our results are consistent with the previous studies.

Visual field assessment is one of the major diagnostic tools in neuroophthalmologic disorders and glaucoma. Perimetric analysis may show involved areas in the visual field even though the patients’ central vision has been preserved. It has been suggested that standard automatic perimetry is a measure of inner retinal ganglion cells (Sehi et al. 2009). Although controversy still exists over the role of blue-yellow perimetry in detecting early visual field defects, SWAP was reported to be more sensitive than standard achromatic perimetry in demonstrating neuroophthalmologic diseases and glaucoma and may show the functional abnormality earlier than the structural one in the optic nerve head or the retina (Keltner & Johnson 1995; Wild 2001; Heijl 2002). Feigl et al. (2011) reported that flicker perimetry could be useful in the detection of ischaemic/hypoxic retinal disorders under photopic and mesopic light levels. In light of these findings, we performed both SAP and SWAP on all study subjects. We found global indices (MD, PSD and CPSD) of both visual field analysis (SAP and SWAP) to be significantly different between both study groups.

Kergoat et al. (2006) studied retinal ganglion cell sensitivity to mild hypoxaemia and showed that ganglion cell function is reduced with decreased arterial blood oxygen. They concluded that neural function during hypoxia is affected as a result of metabolic changes and cannot be compensated properly by vascular regulation of inner retina. Palombi et al. (2006) emphasized the role of reactive oxygen species and related vascular endothelium damage in most respiratory events as a result of desaturation and reoxygenation sequence. Hypoxia-related imbalance between vasoconstrictor endothelin and vasodilatator nitric oxide was assumed to be the cause of ganglion cell death and reduced RNFL thickness in patients with obstructive sleep apnoea syndrome (OSAS) (Kargi et al. 2005). Karakucuk et al. (2008) stated that hypoxia in OSAS is of intermittent character (during sleep), and the visual field defects may be due to optic nerve perfusion defects. In their study, Tsang et al. (2006) studied visual field in moderate-to-severe obstructive sleep apnoea (OSAS) patients and showed a higher incidence of visual field defects and concluded that optic nerve may be damaged directly by anoxic damage or indirectly by decreased optic nerve head blood flow during recurrent apnoeas. The optic nerve damage and retinal ganglion cell death in glaucoma have been explained by increased reactive oxygen species in hypoxic and hypoperfused tissue (Flammer et al. 2002; Ko et al. 2005; Yanagi et al. 2011). Several studies have shown evidence of decreased vasodilation, increased atherosclerosis, arterial stiffness, risk of cardiovascular disease and abnormalities in systemic vascular function in patients with COPD. Eickhoff et al. (2008) reported impaired endothelium-dependent and endothelium-independent vasodilation in patients with stable COPD and did not find a relation between smoking pack years and systemic vascular abnormalities. Mills et al. (2008) found increased arterial stiffness and higher blood pressures in patients with COPD. Endothelial dysfunction and atherosclerotic changes may influence the ocular blood flow and retinal autoregulation (Yanagi et al. 2011). It is known that inner retinal blood supply is autoregulated and the degree of this autoregulation decreases with age (Feigl et al. (2008); Harris et al. 2005). COPD-related vascular changes, low oxygen saturation, increased oxidative stress and ageing might contribute to local ischaemia/hypoxia and may cause a reduction in retinal sensitivity and an increase in MD.

The effects of smoking on eye can be summarized as decreased blood flow in ocular and retinal blood vessels, deranged autoregulation, increased vascular resistance in chorioretinal and optic nerve circulation and abnormal choroidal vascular reactivity (Wimpissinger et al. 2004). The role of increased oxidative stress and lipid peroxidation and reduced oxygen-carrying capacity of haemoglobin has also been emphasized in smokers. Visual field has been studied in smokers. (Hepsen & Evereklioglu 2001; Akarsu et al. 2004). Retinal sensitivity was found to be decreased, and the presence of localized field defects was shown in smokers. In these studies, non-hypoxic young patients with only a history of smoking were included. In this study, we evaluated stable COPD patients with mild–moderate hypoxaemia (mean pO2 = 67.28 ± 7.2 mmHg) (Gilbert & Vender 1995). COPD is a systemic disease and smoking is one of the contributing factors. Twenty-seven of our patients (70.9%) were not a smokers at the time of this study, and we could not detect a correlation between cigarette smoking and parameters of both visual fields (Table 4). Besides, none of our control subjects had a history of smoking (Table 1). Although cigarette smoking is an established factor affecting retinal sensitivity, many other factors (hypoxia, endothelial dysfunction and decreased vascular dilatation) seem to have an effect on COPD-related retinal involvement (Figs 1A,B and 2).

Figure 1.

 (A–B) W-W and B-Y perimetries of a patient with chronic obstructive pulmonary disease showing visual field changes and decreased retinal sensitivity. W-W perimetry shows affected areas in the lower quadrant of the visual field in each eye, and B-Y perimetry demonstrates visual field defect in the lower quadrant of the left eye and more extensive visual field defect in the lower hemifield of the right eye.

Figure 2.

 Normal visual evoked potential recordings of the same patient in each eye.

In conclusion, our study results have suggested that retina and optic nerve seem to be effected by COPD-related hypoxaemia, vascular and endothelial dysfunction. Further studies investigating ocular blood flow, pattern ERG and OCT are needed to explain the retinal and optic nerve involvement in patients with COPD.

Ancillary