Training of vertical versus horizontal reading in patients with hemianopia - a randomized and controlled study

Abstract

Hypothesis: Patients with hemianopic field defects (HFD) might benefit from reading text in vertical orientation if they place the text in the seeing hemifield along the vertical midline.

Methods: We assigned 21 patients with HFD randomly to either vertical or horizontal reading training. They trained reading single lines of texts from a computer screen at home for 2 × 30 min/day, 5 days/week, for 4 weeks. The main outcome variable was reading speed (RS) during reading standardized paragraphs of printed text (IReST) aloud. RS was assessed before training (T1), directly after training (T2) and 4 weeks later (T3). Quality of life (QoL) was assessed by Impact of Visual Impairment (IVI) questionnaire.

Results: Vertical training improved RS in the vertical direction significantly. Only patients with right HFD benefited. Horizontal training improved RS in horizontal diection significantly, but much more in patients with left than in those with right HFD. Both effects remained stable at T3. RS during training at the computer improved highly significantly and correlated strongly with RS of printed text (Pearson r= > 0.9). QoL: Vertical training showed a statistically significant improvement in the complete IVI-score, patients with right HFD in the emotional IVI-score.

Conclusions: The improvements of RS were specific for the training. The stable effect indicates that the patients can apply the newly learned strategies to everyday life. The side of the HFD plays an essential role: Left-HFD patients benefitted from horizontal training, right-HFD patients from vertical training. However, the vertical RS did not reach the level of horizontal RS. The study was registered in the German Clinical Trials register (DRKS-ID: DRKS00018843).

Keywords: Hemianopic field defect; Hemianopic reading disorder; Homonymous hemianopia; Reading training; Text orientation; Vertical reading.

Fig. 1

Visual and perceptual span in normal conditions (top) and in patients with hemianopic field defects (HFD). Top: a) Visual acuity (black) and cone density (blue) depending on eccentricity, the proportions of fovea and foveola (green) and the minimum reading visual field (2° to right and left of fixation and 1° above and below [1]) or visual span or letter recognition span [2]) displayed as turquois oval. b) These data related to a text: Due to the decreasing visual acuity curve (black), the letters are seen clearly only within the turquois oval. The perceptual span during a fixation can be increased up to 5° or 15 letters in reading direction by parafoveal information processing [3]. Bottom: In patients with HFD, reading ability depends on the distance of the field defect to the center: c) in macular splitting, half of the reading visual field is covered by the scotoma and reading is extremely impaired. d) In macular sparing, reading can be preserved. e) Text rotation and vertical reading might be of functional benefit, as patients can shift the text into the seeing hemifield (modified after [8]).

Fig. 2

Study design. The testing times were: T1: before training, T2: testing directly after 4 weeks of training, T3: 4 weeks after end of training and beginning of cross-over training

Fig. 3

Reading speed for single lines of printed text taken from IReST paragraphs in vertical orientation by patients with vertical reading training (group V). 3.1: For the total group: significant improvement of RS from T1 to T2, which remained stable at T3. 3.2: group V separated into subgroups of left or right hemianopic field defect (HFD): No change in left HFD (n = 5), but statistically significant improvement from T1 to T2 in right HFD (n = 6). The effect remained stable at T3. 3.3: Individual change of RS (vertical) during the training period: wide overlap between right (red dots) and left HFD (blue dots). Only three patients improved their reading speed by more than 10 wpm (3.3)

Fig. 4

Reading speed for single lines of printed text taken from IReST paragraphs in horizontal orientation. 4.1: all patients (n = 21): significant improvement of RS from T1 to T2, which remained stable at T3. 4.2: separated by training groups: no change in the vertical training group V, but statistically significant improvement in the horizontal training group H from T1 to T2 and from T1 to T3. The effect remained stable after training. 4.3: Individual change of RS (horizontal): wide overlap between the groups (group V: green dots, group H: orange dots) without a clear advantage for the horizontal training group

Fig. 5

Reading speed for single lines of printed text (IReST) in horizontal orientation. 5.1: total cohort (n = 21) separated by the side of the HFD. There was statistically significant improvement in left HFD (n = 11) from T1 to T2 (p = 0.007), which remained stable. In right HFD (n = 10) there was statistically significant improvement only from T1 to T3 (p = 0.02), and less pronounced than in left HFD. 5.2: Individual improvement of RS (horizontal) during the training period: patients with left HFD (blue dots) also improved more than those with right HFD (red dots). Eight patients with left HFD showed an improvement of more than 10 wpm, only four patients with right HFD

Fig. 6

Reading speed during home training at the computer (cRS) vs. reading speed (RS) during testing with printed text (IReST paragraph) in the clinic. Top: group V, vertical reading training at T1 (left) and T2 (right). Bottom: Group H, horizontal reading training at T1 (left) and T2 (right). cRS and RS in both training groups were strongly correlated (see r-values in graph)