| |

Received: 5 November 2019    Revised: 24 January 2020    Accepted: 9 February 2020

DOI: 10.1111/ane.13232

ORIGINAL ARTICLE

Vision-related quality of life in patients with occipital stroke

Ane Roushan Tharaldsen1  | Kristin Modalsli Sand2,3 | Ingvild Dalen4 |

Gunvor Wilhelmsen5 | Halvor Næss2,6,7 | Anna Midelfart8 | Eyvind Rødahl2,9 |
Lars Thomassen2,7 | Jana Midelfart Hoff7,10 | the NOR-OCCIP Research Group

1
Department of Ophthalmology, Stavanger
University Hospital, Stavanger, Norway Objectives: The aim of this study was to detect visual field defects (VFDs) after oc-
2
Department of Clinical Medicine, University cipital infarction, investigate the rate of recovery and the impact of VFD upon vision-
of Bergen, Bergen, Norway
related quality of life (QoL).
3
Department of Internal Medicine, Sørlandet
Hospital Flekkefjord, Flekkefjord, Norway
Materials and methods: Multicenter, prospective study including patients with MRI
4
Section of Biostatistics, Department of verified acute occipital infarction (NOR-OCCIP project). Ophthalmological examina-
Research, Stavanger University Hospital, tion including perimetry was performed within 2 weeks and after 6 months. Vision-
Stavanger, Norway
5 related QoL was assessed by the National Eye Institute Visual Function Questionnaire
Department of Pedagogy in Teacher
Education, Faculty of Education, Western 25 (VFQ-25) at one and 6 months post-stroke.
Norway University of Applied Sciences,
Results: We included 76 patients, reliable perimetry results were obtained in 66 pa-
Bergen, Norway
6
Centre for Age-Related Medicine, tients (87%) at a median of 8 days after admittance and VFD were found in 52 cases
Stavanger University Hospital, Stavanger, (79%). Evaluation of VFD after 6 months revealed improvement in 52%. Patients with
Norway
7
VFD had significantly lower composite score in VFQ-25 at both test points (77 vs 96,
Department of Neurology, Haukeland
University Hospital, Bergen, Norway P = .001 and 87 vs 97, P = .009), in nine out of eleven subscales of VFQ-25 at 1 month
8
Faculty of Medicine, Norwegian University and seven subscales after 6 months, including mental health, dependency, near and
of Science and Technology, Trondheim,
Norway
distance activities. Milder VFD had better results on VFQ-25 modified composite
9
Department of Ophthalmology, Haukeland score (95 vs 74, P = .002).VFD improvement was related to improved VFQ-25 modi-
University Hospital, Bergen, Norway fied composite score (9.6 vs 0.8, P = .018). About 10% of patients with VFD reported
10
Faculty of Health, VID Specialized
driving 1 month post-stroke and 38% after 6 months.
University, Bergen, Norway
Conclusion: VFD substantially reduces multiple aspects of vision-related QoL.
Correspondence
Severity of VFD is related to QoL and VFD improvement results in better QoL.
Ane Roushan Tharaldsen, Department
of Ophthalmology, Stavanger University Neglecting visual impairment after stroke may result in deterioration of rehabilitation
Hospital, Torgveien 25, 4016 Stavanger,
efforts. Driving post-stroke deserves particular attention.
Norway.
Email: [email protected]
KEYWORDS
Funding information occipital infarction, quality of life, visual field defects
The NOR-OCCIP study is supported by the
Norwegian Directorate of Health.

1 |  I NTRO D U C TI O N there is an increase in the number of stroke survivors living with
residual impairment,3 including visual field defects (VFDs). The re-
Cerebrovascular disease is one of the main causes of disability and ported prevalence of visual field defects after stroke varies consid-
1,2
death in the Western world. With decreasing stroke mortality, erably in the literature. A recent epidemiological study from England

The members of the NOR-OCCIP Research Group are listed in Appendix 1.

© 2020 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Acta Neurol Scand. 2020;141:509–518.  |

wileyonlinelibrary.com/journal/ane     509
 |
510       THARALDSEN et al.

reported a point prevalence of 28% VFD among 1033 patients in Previous stroke or simultaneous acute infarction in other lobes
4
acute stroke units. was not exclusion criteria. Written consent was obtained from all
Occipital infarction is defined as an ischemic stroke in the oc- patients directly or by proxy. Patients with severe cognitive impair-
cipital lobe, caused by a thrombotic or embolic lesion in the distri- ment and/or pre-existing conditions that severely affect eyesight
bution of the posterior cerebral artery, representing 5%-10% of all were excluded.
5
ischemic strokes. Occipital infarctions are primarily associated with In accordance with national guidelines, data related to all stroke
changes in vision, such as hemianopia, which often significantly af- patients are stored electronically in the Norwegian Stroke Register.10
fect daily functions such as driving and reading, reducing personal Extraction of data from this register was used in this study. The
autonomy. Having a visual field defect after ischemic stroke is also NOR-OCCIP study was approved by the Regional Committee for
6
independently associated with increased mortality. medical and health research Ethics in western Norway (2012/2307).
Treatment with thrombolysis after acute infarction is associ-
ated with improved recovery from visual impairment.7 However,
occipital stroke patients often contact healthcare late, their visual 2.2 | Measures
disturbances might be inadequately recognized, and they are often
ineligible for intravenous thrombolysis because they do not arrive Occipital infarction was verified by MRI on admission. The patients
within the therapeutic window for acute intervention.8 This might were examined, diagnosed, and treated in stroke units in the acute
result in a missed opportunity for prevention of permanent visual phase, according to national guidelines.11 Demographic variables,
impairment. lifestyle factors, and information regarding previous diseases and
Although the incidence and point prevalence of visual impair- medication were collected from the Norwegian Stroke Register.
ment in acute stroke are alarmingly high and affect over half of Neurological status was assessed using the National Institutes of
the survivors,4 visual problems are often not reflected in the re- Health Stroke Scale (NIHSS) in the acute phase, and functional out-
habilitation process. The focus has traditionally been on restoring come was measured using modified Rankin Scale (mRS) at 3 months,
motor and speech function. Even if the effect of interventions for both derived from the National Stroke Register.
VFD in stroke has not yet been showed in adequately powered ran- All patients were examined on two occasions by an experienced
domized controlled trials,9 early recognition, support in learning to ophthalmologist using a standardized investigation sheet at base-
cope with the impairment as well as an individualized approach to line (within 2  weeks) and after 6  months. The eye examination in-
develop compensatory functional behavior could greatly benefit cluded best-corrected visual acuity and a general eye examination
the patients. identifying pre-existing ocular pathology. Visual acuity was assessed
The NOR-OCCIP project is a prospective, multicenter, popu- using Snellen or EDTRS eye charts. Assessment of ocular alignment
lation based, representative cohort study with follow-up, using and motility consisted of cover test, evaluation of saccadic, smooth
data from the Norwegian Stroke Register, including patients with pursuit and vergence eye movements, stereopsis (Lang I and II), and
occipital lobe infarction in three university hospitals from August lid function. Visual neglect was assessed by line bisection test, and
2013 to December 2014 (www.clini​
c altr​
ials.gov/ct2/show/ color perception was tested with Ishihara plates.
NCT02​3 07981). Although it is a research study, the NOR-OCCIP The visual field was examined quantitatively by standardized au-
project also has the aim to establish an effective multidisciplinary tomated perimetry using 30-2 threshold tests on most patients. A
diagnostic pathway for patients with VFD and stroke, to be uti- few were tested with Full field 120 point screening. The quality of
lized nationwide. the results was assessed by experienced ophthalmologists. The same
The aims of the present study was to investigate the rate of re- testing strategy was applied for each patient, whereby a direct com-
covery and the impact of visual field defects (VFDs) related to occip- parison could be done. The testing was performed in each eye sep-
ital infarction on vision-related quality of life (QoL). arately using the Octopus 900 perimeter (Haag-Streit Diagnostics)
in Haukeland University Hospital, the Humphrey Automated Field
Analyser II (Carl Zeiss Meditec) in St. Olavs Hospital and Oculus
2 |  M ATE R I A L S A N D M E TH O DS Twinfield 2 perimeter (Oculus) in Stavanger University Hospital. The
perimetry results were classified as: normal, scotoma (<50% of quad-
2.1 | Study design and population rant), quadrantanopia, incomplete hemianopia (<75% of hemifield),
hemianopia, severe bilateral defects.
All patients with acute brain infarction admitted to Haukeland The patients were divided into two main subgroups; non-visual
University Hospital, St. Olavs Hospital and Stavanger University field defect (NVFD) vs VFD. Further subgrouping was mild VFD
Hospital (all in Norway) were considered for inclusion within 7 days (scotoma or quadrantanopia) vs severe VFD (hemianopia or severe
of admittance. The inclusion period was from August 2013 to bilateral defects). Patients with persistent VFD were further divided
December 2014. Patients 18 years of age or above with a permanent into those with improvement vs non-improvement after 6 months.
address in Norway and acute occipital infarction verified by cere- Improvement of VFD after 6 months was defined as ≥3 continuous
bral magnetic resonance imaging (MRI) were included in the study. points improvement on Full field 120 point screening on HFA2 or
THARALDSEN et al. |
      511

mean-deviation improvement of ≥2.0 on 30-2 threshold tests on the

three perimeters compared with baseline results.
NIHSS
The National Eye Institute Visual Function Questionnaire 25 Admission
(VFQ-25) is a global vision-specific health-related quality-of-life MRI brain
questionnaire that consists of 25 items representing 11 subscales
and one single item rating general health. The subscales are gen-
Ophthalmic
eral vision, ocular pain, near vision activities, distance activities, examinaon
social functioning, mental health, role difficulties, dependency, <2 wk
driving, color vision, and peripheral vision. The score produced for
Autoperimetry
the VFQ-25 converts the precoded numeric values of items to a

Occipital stroke
score from 0 to 100. Higher scores reflect better QoL.12 The com-
posite score is calculated by averaging the vision-targeted subscale VFQ-25
scores. Due to many missing observations on the subscale driving, 4 wk
we report a modified composite score excluding driving. The in- Examinaon vision
terviewer-administrated version, translated to Norwegian by Mapi teacher
Research Trust (Lyon, France), was used. Mapi Research Trust did
a linguistic validation process which was officially approved by the 3 mo mRS
developer of the original instrument. The Norwegian version of
VFQ-25 showed satisfactory psychometric properties, including
validity, reliability, responsiveness, discriminatory power, and pre- VFQ-25
dictive value in a population of patients with age-related macu-
lar degeneration.13 VFQ-25 was assessed 4 weeks and 6 months Ophthalmic
after the stroke. The interview was conducted over telephone by examinaon
a study nurse.
6 mo
All patients were additionally examined by a vision teacher, spe- Autoperimetry
cialized in vision, and vision rehabilitation. This was done in accor-
dance with a newly developed national standard, and the data will Examinaon vision
be presented in another publication (Figure 1). teacher

F I G U R E 1   NOR-OCCIP protocol. NIHSS, the National Institutes

of Health Stroke Scale; MRI, magnetic resonance imaging; VFQ-25,
2.3 | Statistical analysis the National Eye Institute Visual Function Questionnaire 25; mRS,
modified Rankin Scale
Descriptive statistics are presented as counts and percentages for
categorical data and as medians and interquartile ranges (IQR) for
continuous data. Comparisons of categorical variables between 3 | R E S U LT S
groups were performed using Pearson's chi-squared test or Fisher's
exact test. Due to skewed distributions, comparisons of continuous A total of 76 patients with acute occipital infarction were included;
data between groups were performed with Mann-Whitney U test. For 31 patients from Haukeland University Hospital, 21 patients from St.
adjusted analysis, the VFQ-25 subscale scores were first reversed, Olavs Hospital, and 24 patients from Stavanger University Hospital.
then log-transformed to improve the symmetry of the distributions. The total number of stroke patients from these three hospitals
Hence, adjusted between-groups comparisons of QoL dimensions registered in the inclusion period was 1969 patients (Haukeland
were performed using linear regression analysis with these trans- University Hospital: 528 patients, St.Olavs Hospital: 674 patients
formed VFQ-25 scores as dependent variables. For between-groups and Stavanger University Hospital: 767 patients). Thus, the rate of
comparisons of changes in QoL dimensions, the change scores are occipital infarctions in general stroke was 4%.
presented as means and standard deviations (SD), and we used inde- Confrontation visual field testing as a part of NIHSS at admis-
pendent samples t tests for unadjusted comparisons, applying Welch sion showed VFD in 47 of 75 patients (63%) at admission (one miss-
correction if within groups SDs were dissimilar; and adjusted compar- ing). Complete ophthalmological examination (including perimetry)
isons were performed using linear regression analysis with changes within 2 weeks after admittance was obtained in 66 patients (87%),
scores as dependent variables. All analyses were based on all avail- three patients died and seven did not show up (Figure 2). The median
able cases for each particular analysis and were performed within the (IQR) time from admittance was 8 (4-12) days to the first ophthalmic
statistical package SPSS version 23. Estimated P-values < .05 were examination (n = 66) and 191 (168-232) days to the follow-up exam-
considered statistically significant. ination (n = 57).
 |
512       THARALDSEN et al.

F I G U R E 2   Inclusion NOR-OCCIP
project

The demographic data of the 66 patients with complete ophthal- examination were similar in patients with non/mild/severe VFD
mological examination in the acute phase are presented in Table 1. (14%/18%/20%, P = .91).
The median (IQR) age was 69.7 years (57.4-77.7), and the majority In all, 62 of the total 76 patients (82%) in this study participated
of patients (68%) were men. Pre-existing hypertension was the in the first assessment of VFQ-25 4 weeks post-stroke (Table 2).
most common (52%) comorbidity, followed by hypercholesterolemia Patients with VFD had significantly lower scores regarding total vi-
(37%). Previous stroke/TIA was reported in 25%. Forty percent of sion-specific QoL; as seen on composite score of median (IQR) 77
the patients had a history of smoking. The NIHSS scores were gener- (62, 88) vs 96 (91, 99), P = .001 and modified composite score of
ally low, but higher in the VFD group at baseline compared with the median (IQR) 84 (67, 94) vs 97 (90, 100), P < .001. Subscales with sig-
NVFD group, median (IQR) 2.0 (1.0, 4.0) vs 0.0 (0.0, 1.0), P = .001, as nificant differences between patients with VFD vs patients without
were mRS scores after 3 months, median (IQR) 2.0 (1.0, 2.0) vs 1.0 VFD included the following (adjusted for age and sex): General vision
(0.0, 1.0), P = .001. As for age, sex, or premorbidity, there were no (P < .001), near activities (P = .04), distance activities (P = .035), so-
significant differences between the groups. cial functioning (P = .024), mental health (P = .001), role difficulties
At the first ophthalmological examination, 52 of the 66 patients (P = .021), dependency (P = .024), driving (P = .001), and peripheral
(79%) had VFD. In these 52 patients, homonymous hemianopia was vision (P = .016). Also when looking at reading separately, we found
found in 26 (50%), severe bilateral VFD in 15 (29%), quadrantanopia that patients with VFD were significantly more affected with median
in six (11%), and scotoma in five (10%) of the patients. In 14 (21%) (IQR) scores of 75 (25, 100) vs 100 (100, 100), unadjusted P = .005,
of the 66 patients tested with perimetry, no defect was revealed. adjusted P = .003.
The VFDs were right-sided in 40% of the patients, left-sided in 31%, After 6 months, 55 of the 62 patients (89%) participated in a new
while 29% of the patients had visual field defects in both hemifields. assessment of VFQ-25 (Table 2). Patients with VFD at the first oph-
After 6 months, 54 of the 66 patients (82%) attended the fol- thalmological examination continued to score significantly lower on
low-up examination where perimetry showed VFD in 42 (78%). vision-specific QoL demonstrated by the composite score of median
Among these 42, homonymous hemianopia was found in 16 (38%), (IQR) 87 (72, 93) vs 97 (91, 100), P = .009 and modified compos-
scotoma in 14 (33%), severe bilateral VFD in seven (17%), and qua- ite score of median (IQR) 91 (75, 95) vs 95 (92, 100), P = .011. The
drantanopia in five (12%). Two patients with hemianopia at baseline following subscales remained significantly different between pa-
had full recovery after 6 months. tients with VFD vs patients without VFD (adjusted for age and sex):
In the acute phase, 79% (41/52) of the patients with VFD were General vision (P = .013), near activities (P = .019), distance activities
classified as severe (hemianopia or severe bilateral VFD) and 21% (P = .015), mental health (P = .011), role difficulties (P = .026), depen-
(11/52) as mild (scotoma or quadrantanopia). After 6 months, the fre- dency (P = .047), and driving (P = .017). Social functioning (P = .12)
quency of severe VFD declined to 55% (23/42). Improvement of VFD and peripheral vision (P = .18) had improved.
was registered in 22 of the 42 with follow-up perimetry, that is, 52%. At the direct question whether they were currently driving
There was no significant difference in improvement between those or not, 10% with VFD replied they were driving after 1 month
with mild vs severe VFD in the acute phase (44% vs 55%, P = .71, (n = 49) and 38% after 6 months (n = 37). Of those driving with VFD
Fisher's exact test). Dropout rates after the first ophthalmological after 6 months, nine patients had scotoma, two had incomplete
 THARALDSEN et al.

TA B L E 1   Demographic data on 66 patients with occipital infarction, by visual field defects at baseline

VFD

  Valid n NVFD/VFD All (n = 66) NVFD (n = 14) Total (n = 52) Mild (n = 11) Severe (n = 41) P* P**

Age (y), median (IQR) 14/52 69.7 (57.4, 77.7) 66.7 (52.9, 71.5) 72.1 (59.8, 78.9) 71.5 (53.6, 78.0) 72.4 (61.4, 79.3) .054 .52
Females 14/52 21 (32%) 2 (14%) 19 (37%) 3 (27%) 16 (39%) .11 .47
Hypertension 14/51 34 (52%) 5 (36%) 29 (57%) 4 (36%) 25 (63%) .16 .12
Hypercholesterolemia 14/51 24 (37%) 5 (36%) 19 (37%) 4 (36%) 15 (38%) .92 .95
Atrial fibrillation 14/51 15 (23%) 1 (7%) 14 (28%) 3 (27%) 11 (28%) .11 .99
Previous myocardial 14/51 11 (17%) 2 (14%) 9 (18%) 4 (36%) 5 (13%) .77 .066
infarction
Previous stroke 14/51 11 (17%) 4 (29%) 7 (14%) 0 (0%) 7 (18%) .19 .14
Previous TIA 14/50 5 (8%) 1 (7%) 4 (8%) 1 (9%) 3 (8%) 1.0 1.0
Diabetes 14/51 5 (8.0%) 1 (7%) 4 (8%) 0 (0%) 4 (10%) 1.0 .57
Cigarette smoking 14/48 25 (40%) 5 (36%) 20 (42%) 6 (60%) 14 (37%) .69 .19
NIHSS score, median (IQR) 13/47 2.0 (0.0, 4.0) 0.0 (0.0, 1.0) 2.0 (1.0, 4.0) 1.0 (0.0, 3.0) 3.0 (1.0, 4.3) .001 .048
mRS score 3 mo, median 10/41 2.0 (1.0, 2.0) 1.0 (0.0, 1.0) 2.0 (1.0, 2.0) 2.0 (0.0, 2.0) 2.0 (1.0, 3.0) .001 .15
(IQR)
BCVA (LogMAR) > 0.3 14/52 3 (5%) 0 (0%) 3 (6%) 0 (0%) 3 (7%) 1.0 1.000

Note: Statistics given as counts (percentages) unless otherwise stated.

Abbreviations: BCVA, best-corrected visual acuity; IQR, interquartile range; mRS, modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; NVFD, non-visual field defect; TIA, transient
ischemic attack; VFD, visual field defect.
*P-value from comparison NVFD vs VFD using Mann-Whitney U test or Pearson's chi-squared test as appropriate.
**P-value from comparison mild vs severe VFD using Mann-Whitney U test, Pearson's chi-squared test, or Fisher's exact test as appropriate.
Bold values are indicates p < .05.
|
      513
 |
514      

TA B L E 2   Baseline and follow-up VFQ-25 subscales by baseline visual field defect among 62 patients with occipital infarction

VFQ-25 within 4 wk post-stroke VFQ-25 at 6 mo post-stroke

NVFD (n = 12) VFD (n = 50) NVFD (n = 13) VFD (n = 42)

  Valid n NVFD/VFD Median (IQR) Median (IQR) P* P** Valid n NVFD/VFD Median (IQR) Median (IQR) P* P**

Composite score 11/29 96 (91, 99) 77 (62, 88) .001 .001 12/28 97 (91, 100) 87 (72, 93) .006 .009
Modified composite 12/46 97 (90, 100) 84 (67, 94) .001 <.001 13/41 95 (92, 100) 91 (75, 95) .006 .011
score† 
General health 11/50 50 (50, 75) 50 (25, 56) .15 .87 13/42 50 (38, 75) 50 (25, 75) .29 .75
General vision 12/50 90 (80, 100) 6 (40, 80) <.001 <.001 13/42 80 (80, 100) 70 (60, 80) .006 .013
Ocular pain 12/50 100 (88, 100) 100 (84, 100) .94 .89 13/42 100 (75, 100) 100 (88, 100) .99 .78
Near activities 12/50 100 (94, 100) 75 (50, 100) .005 .004 13/42 100 (92, 100) 83 (65, 100) .011 .019
Distance activities 12/48 100 (89, 100) 83 (53, 100) .023 .035 13/42 100 (100, 100) 92 (75, 100) .003 .015
VS Social functioning 12/49 100 (100, 100) 100 (88, 100) .031 .024 13/41 100 (100, 100) 100 (100, 100) .089 .12
VS Mental health 12/50 97 (89, 100) 81 (38, 94) .002 .001 13/42 100 (94, 100) 88 (75, 100) .006 .011
VS Role difficulties 12/49 100 (81, 100) 75 (25, 100) .011 .021 13/42 100 (82, 100) 81 (34, 100) .027 .026
VS Dependency 12/49 100 (100, 100) 100 (67, 100) .015 .024 13/42 100 (100, 100) 100 (92, 100) .018 .047
Driving 11/30 92 (0, 100) 0 (0, 0) .001 .001 12/28 100 (89, 100) 42 (0, 100) .008 .017
Color vision 12/49 100 (100, 100) 100 (100, 100) .23 .29 13/41 100 (100, 100) 100 (100, 100) .70 .50
Peripheral vision 12/46 100 (100, 100) 75 (44, 100) .018 .016 13/42 100 (100, 100) 100 (69, 100) .11 .18

Abbreviations: IQR, interquartile range; NVFD, non-visual field defect; VFD, visual field defect; VFQ-25, National Eye Institute Visual Functioning Questionnaire-25; VS, vision specific.
*P-value from Mann-Whitney U test.
**P-value from linear regression with log-transformed reversed VFQ-25 subscale score as dependent variable, VFD vs NVFD as independent variable, and adjusted for age and sex.
†
Composite score excluding driving.
Bold values are indicates p < .05.
THARALDSEN et al.
THARALDSEN et al. |
      515

homonymous hemianopia and three had severe, bilateral VFD tested with significantly larger improvement after adjustment included:
with perimetry. General health (P = .022), general vision (P = .001), distance activities
The VFDs of the 62 patients participating in the first assessment (P = .007), and social functioning (P = .031). The difference in im-
of VFQ-25 were stratified further by severity (Table 3). Patients with provement of the composite score was not significant after adjusting
mild VFD vs severe VFD scored better on their total vision-specific (P = .097), neither were the differences in improvement of ocular
QoL defined by modified composite score of median (IQR) 95 (89, pain, near activities, mental health, role difficulties, dependency,
97) vs 74 (63, 89), P = .002, but the total composite score was not driving, color vision, or peripheral vision.
significantly different. Subscales with significantly difference re- Among the 76 patients included, 15 (20%) were treated with in-
lated to severity (adjusted for age and sex) included: General vision travenous thrombolysis in the acute phase. One of these patients
(P = .001), near activities (P = .021), mental health (P = .005), role was also treated additionally with thrombectomy. Among the pa-
difficulties (P = .013), dependency (P = .044), and peripheral vision tients who received thrombolysis, severe VFD was found in ten pa-
(P = .002). No significant differences were found for general health, tients and mild VFD in one at the first ophthalmological examination.
ocular pain, distance activities, social functioning, driving, and color
vision (Table 3).
A total of 36 of the 52 patients (69%) who had VFD at the oph- 4 | D I S CU S S I O N
thalmological examination in the acute phase, completed both the
follow-up perimetry and VFQ-25 (Table 4). Of these 36 patients, The main focus of this study was on patients with occipital brain
improvement in VFD was seen in 19 patients (53%). Patients with infarction, their visual impairment and the effect of this upon their
VFD improvement after 6  months also experienced a significantly quality of life. In this study, the majority (63%) of the patients with
larger improvement in vision-specific QoL than those without VFD occipital stroke displayed a VFD at admittance. They were tested
improvement, with change in modified composite score of 9.6 vs 0.8, with confrontational testing as a part of NIHSS, which has been
amounting to a difference (95% CI) of 8.8 (2.1-15.6), P = .012 and an proven to underestimate the frequency of stroke-related VFD.14
age and sex-adjusted difference of 9.2 (1.7-16.7), P = .018. Subscales Full visual assessment including perimetry was done at a median of

TA B L E 3   Baseline VFQ-25 subscales

VFQ-25 within 4 wk post-stroke
by baseline visual field defect among
50 patients with occipital infarction, Severe VFD
comparison mild vs severe VFD Mild VFD (n = 11) (n = 39)
Valid n mild/
  severe VFD Median (IQR) Median (IQR) P* P**

Composite score 7/22 88 (78, 95) 72 (60, 84) .098 .11

Modified 11/35 95 (89, 97) 74 (63, 89) .002 .002
composite score† 
General health 11/39 50 (25, 75) 50 (25, 50) .34 .43
General vision 11/39 80 (80, 100) 60 (40, 80) .001 .001
Ocular pain 11/39 100 (100, 100) 100 (75, 100) .48 .50
Near activities 11/39 100 (83, 100) 67 (50, 100) .024 .021
Distance activities 11/37 92 (75, 100) 83 (50, 100) .17 .19
VS Social 11/38 100 (100, 100) 100 (75, 100) .10 .080
functioning
VS Mental health 11/39 94 (81, 100) 75 (38, 88) .004 .005
VS Role difficulties 11/38 100 (88, 100) 50 (25, 100) .011 .013
VS Dependency 11/38 100 (100, 100) 92 (50, 100) .035 .044
Driving 7/23 0 (0, 100) 0 (0, 0) .22 .27
Color vision 11/38 100 (100, 100) 100 (75, 100) .046 .051
Peripheral vision 11/35 100 (100, 100) 75 (25, 100) .004 .002

Abbreviations: IQR, interquartile range; VFD, visual field defect; VFQ-25, National Eye Institute
Visual Function Questionnaire -25; VS, vision specific.
*P-value from Mann-Whitney U test, mild VFD vs severe VFD.
**P-value from linear regression with log-transformed reversed VFQ-25 subscale score as
dependent variable, mild VFD vs severe VFD as independent variable, adjusted for age and sex.
†
Composite score excluding driving.
Bold values are indicates p < .05.
516       | THARALDSEN et al.

TA B L E 4   Changes in VFQ-25 subscales in patients with occipital infarction from 4 wk to 6 mo post-stroke for 36 patients with initial
VFD, stratified by improvement (yes/no)

Valid n Improved VFD Not improved VFD Unadjusted Adjusted

Improved VFD
  yes/no Mean (SD) Mean (SD) Diff. (95% CI) P* Diff. (95% CI) P**

Composite score 10/10 11.6 (8.5) 3.1 (7.6) 8.5 (0.9, 16.1) .031 6.2 (−1.2, 13.6) .097
Modified composite 19/16 9.6 (12.6) 0.8 (6.3) 8.8 (2.1, 15.6) .012 9.2 (1.7-16.7) .018
score† 
General health 19/17 13.2 (30.5) -8.8 (23.3) 22.0 (3.4, 40.5) .022 23.0 (3.5, 42.6) .022
General vision 19/17 10.5 (15.4) -8.2 (12.4) 18.8 (9.2, 28.3) <.001 18.2 (8.3, 28.2) .001
Ocular pain 19/17 5.3 (22.9) -0.7 (22.7) 6.0 (−9.5, 21.5) .44 4.5 (−11.7, 20.8) .57
Near activities 19/17 8.3 (19.6) -2.0 (18.8) 10.3 (−2.7, 23.3) .12 10.8 (−2.9, 24.5) .12
Distance activities 19/17 10.5 (24.5) -11.8 (24.5) 22.3 (5.7, 38.9) .010 24.2 (7.1, 41.4) .007
VS Social 19/16 7.2 (15.2) -7.8 (18.2) 15.0 (3.6, 26.5) .012 13.2 (1.3, 25.1) .031
functioning
VS Mental health 19/17 11.5 (19.2) 6.3 (16.4) 5.3 (−6.9, 17.4) .39 5.7 (−7.2, 18.5) .38
VS Role difficulties 19/17 5.3 (29.0) 7.4 (24.2) -2.1 (−20.3, 16.1) .82 0.5 (−18.4, 19.3) .96
VS Dependency 19/17 7.5 (18.0) 5.9 (10.1) 1.6 (−8.5, 11.6) .75 2.5 (−8.0, 13.0) .68
Driving 10/10 43.3 (47.9) 25.0 (41.6) 18.3 (−23.8, 60.5) .37 6.7 (−35.9, 49.4) .74
Color vision 19/16 10.5 (22.5) 4.7 (13.6) 5.8 (−6.8, 18.5) .35 7.8 (−5.9, 21.4) .26
Peripheral vision 19/16 19.7 (24.4) 9.4 (22.1) 10.4 (−5.8, 26.5) .20 9.9 (−7.3, 27.1) .25

Abbreviations: CI, confidence interval; Diff., difference; SD, standard deviation; VFD, visual field defect; VFQ-25, National Eye Institute Visual
Function Questionnaire 25; VS, vision specific.
*P-value from independent samples t test, with Welch correction when appropriate.
**P-value from linear regression with change in VFQ-25 subscale score as dependent variable, improved vs not improved as independent variable,
and adjusted for age and sex.
†
Composite score excluding driving.
Bold values are indicates p < .05.

8 days, which is somewhat earlier than in other comparable stud- all stroke patients.17 This may be attributed to recent patient educa-
4,15
ies. With perimetry, the frequency of detected VFD in our study tion programs, both among the public and health caretakers, where
was higher at admittance, 79%. Our numbers are in accordance with the association between a VFD and stroke has been highlighted.
a large prospective multicenter trial where VFD was found in 52% of However, it is worth noticing that the majority of patients with VFD
a general stroke population,16 while 86% VFD was reported among who received thrombolysis in our study were severely affected. This
occipital stroke patients with mainly visual symptoms admitted to an could have made the selection for acute treatment easier.
emergency department at a university hospital in Finland.8 Being familiar with the recovery pattern is important in patient
The mean age of the patients was about 70 years, two thirds education, contact, and follow-up as well as in evaluating the results
were men, and hypertension (52%) and smoking (40%) were the of rehabilitation. In this study, 21% of those who underwent ophthal-
most frequent risk factors. The demographics were similar to those mological examination in the acute phase did not display a VFD. This
presented in the recent study of occipital stroke in Finland, yet the could be attributed to rapid recovery in some of the cases. Other
frequency of hypertension, hyperlipidemia, and diabetes was lower studies have suggested that approximately 19% of post-stroke pa-
in our population.8 However, the rate of smoking in our patients was tients with total homonymous hemianopia experience complete re-
substantially higher than the 18% found in the general stroke popu- covery within 1 month.7
lation in Norway.17 Many patients improve, although they do not fully recover. Fifty-
Approximately 5%-10% of all strokes are posterior cerebral ar- two percent of the patients in our study experienced improvement
tery strokes.5 Many of these patients display visual symptoms, such of their VFD during the first 6 months. This is in accordance with an-
as VFD. However, both patients and healthcare professionals may other study, where they found some improvement of the visual field
have difficulties in recognizing visual problems as related to acute in 50%-60% of patients with homonymous hemianopia within the
stroke, and the thrombolysis rate in patients with occipital infarc- first month.18 Thereafter, recovery seems to decrease with increas-
8
tion has been reported as low as 6.5%. This was not the case in our ing time from the injury, where most of the improvement is expected
study, where the thrombolysis rate was 20%. This is fully in accor- to occur within the first 2 months.18 On the other hand, detailed sta-
dance with the Norwegian national rate of 21% for thrombolysis in tistical analysis of visual recovery in 32 patients with occipital lobe
THARALDSEN et al. |
      517

infarction examined with perimetry demonstrated an improvement remains complicated in the patient-physician relationship, but must
19
in all selected visual field zones after 6 months. be addressed in order to avoid accidents and serious injuries. We
In this study, 79% of the VFDs in the acute phase were classified recommend that stroke units include the issue of driving in their rou-
as severe. At the follow-up perimetry 6 months later, the frequency tines when releasing patients from the unit.
of severe VFD had declined to 55%. This illustrates a clear shift to- The strengths of this study are its multidisciplinary approach
ward milder VFD in our patients. By using VFQ-25, we found that and thorough mapping of the patients, and its relatively long fol-
patients with severe VFD were more prone to have reduced mental low-up time. The early and full visual assessment of the patients is
health as well as problems with dependency and role difficulty. The also a strength, as it could lead to early intervention, with a posi-
severity of VFD has an impact on quality of life, and we argue that tive impact on the rehabilitation outcome. The limitations of the
this should be taken into account when deciding upon whether to study are a large amount of tests and a relatively small sample size
give acute treatment (thrombolysis/thrombectomy) in patients with of occipital stroke patients. In addition, all patients were recruited
VFD. The lesser the injury, the better the outcome. from three university hospitals, which offer full-scale services on
The vast majority of the patients (78%) in this study had a per- the highest national level. This could be more difficult to obtain in
sistent VFD 6 months after stroke, mild, or severe. We live in a visual smaller hospitals.
world, where our ability to register visual input is constantly in de- In conclusion, this study found that 4/5 patients with occipital in-
mand. Reduced visual function does not only affect visual abilities farction had detectable visual field defects. Improvement was found
such as reading, but also social interactions with others and personal in over 50% on 6  months follow-up perimetry. Patients with VFD
20,21
independence. As shown in our study, there was a general im- had significantly lower scores on total vision-related QoL, as well
provement in VFQ-25 in many parameters among patients both with as on several subscales at baseline and after 6 months compared
and without VFD during the 6-month period after stroke, but most to patients without VFD. Driving difficulties were frequent among
differences between these two groups continued to be significant. the patients; however, almost 4/10 patients with VFD were driving
The general improvement in VFQ-25 could be attributed to coping 6 months post-stroke. In our opinion, early identification of visual
mechanisms and general adjustment to the situation after the acute problems in acute stroke patients would benefit patients, caregiv-
phase. ers, and the healthcare system in general. Optimal multidisciplinary
Mental health is an important factor related to coping with im- structures for this should be established, as also suggested by oth-
pairment and quality of life. When evaluating the follow-up of VFQ- erss.4,24 Stroke-related visual field defects are a silent, but serious
25, we found that patients who had VFD continued to struggle with handicap, easily overlooked by healthcare workers, patients, and
mental health, regardless if their VFD had improved or not. Having surroundings.
a VFD after stroke has previously been found to be independently
associated with higher mortality.6 On the basis of our findings, we AC K N OW L E D G M E N T S
postulate that having a VFD is also independently associated with The authors have no acknowledgment to declare.
reduced mental health, although other factors such as stroke-related
depression and cognitive impairment undoubtedly can contribute to C O N FL I C T O F I N T E R E S T
this health worsening. The authors report no conflicts of interest in this work.
Driving is the primary mode of transportation in many Western
countries and therefore an important factor in personal indepen- DATA AVA I L A B I L I T Y S TAT E M E N T
dence and integrity. Cessation of driving is associated with higher The anonymized data that support the findings of this study are
22
levels of depression. One of the critical issues in traffic ophthal- available on request from the first author, ART.
mology is driving with visual field defects, as most of the informa-
tion regarding driving is visual. We found that even if the patients’ ORCID
VFD improved, they still found driving difficult. In Norway, patients Ane Roushan Tharaldsen  https://orcid.
are restricted from driving when field defects are overlapping in the org/0000-0001-7808-7992
central 20 degrees (binocular testing). The peripheral visual field
must be at least 120 degrees in the horizontal plane, 50 degrees to REFERENCES
each side and 20 degrees in the vertical plane. This implies that most 1. Feigin VL, Forouzanfar MH, Krishnamurthi R, et al. Global and
of the patients in the VFD group, except those with a small periph- regional burden of stroke during 1990–2010: findings from the
Global Burden of Disease Study 2010. Lancet (London, England).
eral defect, did not meet the criteria for driving.
2014;383(9913):245-254.
To our surprise, we found that about 10% of the patients with 2. Lopez AD, Mathers CD, Ezzati M, Jamison DT, Murray CJ. Global
VFD reported driving 4 weeks post-stroke and 38% after 6 months. and regional burden of disease and risk factors, 2001: system-
Our results are in accordance with a former study that also report on atic analysis of population health data. Lancet (London, England).
2006;367(9524):1747-1757.
post-stroke drivers with significantly affected visual fields.14 At the
3. Guzik A, Bushnell C. Stroke epidemiology and risk factor manage-
same time, about 5% of the eldest group of European drivers have vi- ment. Continuum (Minneapolis, Minn). 2017;23(1, Cerebrovascular
sual field either borderline or below standard. 23 The issue of driving Disease):15-39.
 |
518       THARALDSEN et al.

4. Rowe FJ, Hepworth LR, Howard C, Hanna KL, Cheyne CP, Currie 21. Papageorgiou E, Hardiess G, Schaeffel F, et al. Assessment of vi-
J. High incidence and prevalence of visual problems after acute sion-related quality of life in patients with homonymous visual
stroke: an epidemiology study with implications for service deliv- field defects. Graefe's archive for clinical and experimental ophthal-
ery. PloS one. 2019;14(3):e0213035. mology = Albrecht von Graefes Archiv fur klinische und experimentelle.
5. Brandt T, Steinke W, Thie A, Pessin MS, Caplan LR. Posterior cere- Ophthalmologie. 2007;245(12):1749-1758.
bral artery territory infarcts: clinical features, infarct topography, 22. Ragland DR, Satariano WA, MacLeod KE. Driving cessation and
causes and outcome. Multicenter results and a review of the litera- increased depressive symptoms. J Gerontol Ser A, Biol Sci Med Sci.
ture. Cerebrovasc Dis (Basel, Switzerland). 2000;10(3):170-182. 2005;60(3):399-403.
6. Sand KM, Naess H, Thomassen L, Hoff JM. Visual field defect 23. van Rijn LJ, Nischler C, Michael R, et al. Prevalence of impair-
after ischemic stroke-impact on mortality. Acta Neurol Scand. ment of visual function in European drivers. Acta Ophthalmol.
2018;137(3):293-298. 2011;89(2):124-131.
7. Ali M, Hazelton C, Lyden P, Pollock A, Brady M. Recovery from 24. Goodwin D. Homonymous hemianopia: challenges and solutions.
poststroke visual impairment: evidence from a clinical trials re- Clinical Ophthalmol (Auckland, NZ). 2014;8:1919-1927.
source. Neurorehabil Neural Repair. 2013;27(2):133-141.
8. Raty S, Silvennoinen K, Tatlisumak T. Prehospital pathways of oc-
cipital stroke patients with mainly visual symptoms. Acta Neurol
How to cite this article: Tharaldsen AR, Sand KM, Dalen I, et al.
Scand. 2018;137(1):51-58.
9. Hazelton C, Campbell P, Rowe FJ, et al. Interventions for visual field
Vision-related quality of life in patients with occipital stroke.
defects in people with stroke. Stroke. 2019;50(12):e419-e420. Acta Neurol Scand. 2020;141:509–518. https://doi.
10. Norwegian Stroke Register. Report 2013. https://www.kvali​tetsr​ org/10.1111/ane.13232
egist​re.no/regis​ters/norsk-hjern​eslag​register. Accessed 23 January,
2020.
11. National guidelines for treatment and rehabilitation for ischemic
stroke. Norwegian Directorate of Health 2010, IS-1688, http://
www.helse​d irek ​torat​e t.no/publi​k asjo​n er/nasjo​n al-retni​n gsli​n je- APPENDIX 1
for-behan​dling-og-rehab​ilite​ring-ved-hjern​eslag-fullv​ersjo​n/Sider/​
defau​lt.aspx T H E N O R- O C C I P R E S E A R C H G R O U P
12. Mangione CM, Lee PP, Gutierrez PR, Spritzer K, Berry S, Hays
Leila Marie Frid (Study Nurse, Haukeland University Hospital,
RD. Development of the 25-item national eye institute visual
function questionnaire. Arch Ophthalmol (Chicago, Ill : 1960).
Bergen, Norway); Hilde Karine Tandstad (Vision Teacher, Statped
2001;119(7):1050-1058. Vest, Bergen, Norway); Grethe Hegreberg (Vision Teacher,
13. Jelin E, Wisloff T, Moe MC, Heiberg T. Psychometric properties Stavanger municipality, Norway); Kristin Lundberg (Vision Teacher,
of the National Eye Institute Visual Function Questionnaire (NEI- Trondheim municipality, Norway); Tom Rune Karlsen (Faculty
VFQ 25) in a Norwegian population of patients with neovascular
of Medicine, Norwegian University of Science and Technology,
age-related macular degeneration compared to a control popula-
tion. Health Quality Life Outcomes. 2019;17(1):140. Trondheim, Norway); Bjørn Setseng (Faculty of Medicine,
14. Townend BS, Sturm JW, Petsoglou C, O'Leary B, Whyte S, Crimmins Norwegian University of Science and Technology, Trondheim,
D. Perimetric homonymous visual field loss post-stroke. J Clin Norway); Gitta Rohweder (Department of Medicine, St Olavs
Neurosci. 2007;14(8):754-756.
Hospital, Trondheim University Hospital, Norway. Department of
15. Rowe F. Symptoms of stroke-related visual impairment. Strabismus.
2013;21(2):150-154. Neuromedicine and Movement Science, Norwegian University of
16. Rowe FJ, Wright D, Brand D, et al. A prospective profile of visual Science and Technology, Norway); Bent Indredavik (Department
field loss following stroke: prevalence, type, rehabilitation, and out- of Medicine, St Olavs Hospital, Trondheim University Hospital,
come. BioMed Res Int. 2013;2013:719096.
Norway. Department of Neuromedicine and Movement Science,
17. Norwegian Stroke Register. Report 2018. https://www.kvali​tetsr​
egist ​ r e.no/sites/​d efau ​ l t/files/​ 1 _arsra ​ p port_2018_hjern ​ e slag_ Norwegian University of Science and Technology, Norway); Martin
des19.pdf. Accessed 23 January, 2020. W. Kurz (Department of Neurology, Stavanger University Hospital,
18. Zhang X, Kedar S, Lynn MJ, Newman NJ, Biousse V. Natural history Norway; Neuroscience Research Group, Stavanger University
of homonymous hemianopia. Neurology. 2006;66(6):901-905.
Hospital, Norway; Department of Clinical Medicine, University of
19. Celebisoy M, Celebisoy N, Bayam E, Kose T. Recovery of visual-field
defects after occipital lobe infarction: a perimetric study. J Neurol Bergen, Norway); Titto Idicula (Department of Neurology, St Olavs
Neurosurg Psychiatry. 2011;82(6):695-702. Hospital, Trondheim, Norway).
20. Chen CS, Lee AW, Clarke G, et al. Vision-related quality of life in
patients with complete homonymous hemianopia post stroke. Top
Stroke Rehabil. 2009;16(6):445-453.