2 year follow-up data was available for 371/460 (81%) of participants, without significant differences in NCS and age at enrolment or treatment within 2 years, between those included and those lost to follow-up (Figure ).
Participation by treatment, received within 2 years of diagnosis.
Of the 371 participants in the study, 195 (53%) had no treatment within two years, 63 (17%) had treatment for reduced visual acuity only, 50 (13%) had non-surgical treatment for X(T) whilst 63 (17%) participants had surgery within two years of enrolment. Figure provides a breakdown of the types of treatment, and detail of participation by treatment received within 2 years. Only 2 children (1 male, 1 female) had constant exotropia at 2 year follow-up. These children are included in the main analysis. 13/63 (21%) of the surgical group had a persistent overcorrection at 6 months post-surgery.
Newcastle Control Score (NCS)
The mean NCS at enrolment was higher (worse) in the surgical group (4.98) compared to the observation (3.43), vision only (3.38) and non-surgical treatment (4.02) groups (Table ). At follow-up the total mean NCS in the surgery group was significantly reduced, on average more than 60% in both the parental and clinic component scores. There were significant but smaller reductions in the total mean NCS at follow-up in the non-surgical treatment and observation groups reflecting an average 20% reduction in the clinical component and 13% reduction in the parental component respectively.
Change in mean and median total NCS by treatment group
Changes in NCS from enrolment to follow-up are shown by treatment group in Table . Overall in the observation group, 19 (10%) improved, 163 (85%) stayed stable and 9 (5%) deteriorated; in the non-surgical treatment group 7 (15%) improved, 39 (81%) stayed stable and 2 (4%) deteriorated; in the treatment for vision only group, 5 (8%) improved, 53 (84%) stayed stable and 5 (8%) deteriorated; in the surgery group 26/62 (42%) improved, 21 (34%) stayed stable, 2 (3%) deteriorated while 13 (21%) were overcorrected.
Change in total Newcastle Control Score by treatment group
A score at final outcome of 0 (no detectable strabismus) was found in 15/62 (24%) of the surgery group plus a further 4/62 (6%) who received Botulinum toxin to treat overcorrection within 6 months of surgery, compared with 5/191 (2.5%) in the observation group, 1/48 (2%) in the non-surgical treatment group and 4/63 (6%) in the group treated only for reduced visual acuity.
There was no statistically significant difference in overcorrection rates by type of surgery (7/27 (26%) in the bilateral compared to 6/36 (17%) in the unilateral group; Chi-Square test 0.80, p = 0.37). Of the 13 overcorrections, mean [SD] total NCS, home control and clinic control at enrolment were 5.08 [1.71], 2.08 [0.86] and 3.0 [1.29] respectively (ranges 2 to 8, 1 to 3 and 1 to 6 respectively).
Changes in angle of X(T)
Table provides details of near and distance angle. At enrolment the median angle was 14 diopters for near (range -6 to 50) and 25 diopters for distance (range 10 to 50) in the non-surgical groups and did not vary over time. In the surgery group the median angle at enrolment was 16 for near (range 1 to 45) and 30 for distance (range 20 to 60), and there were clinically significant changes at follow-up: near angle was reduced from 16 to 7, and distance angle from 30 to 10 (Table ).
Change in near and distance angle by treatment group
212 participants were aged 4 years or older at enrolment, of whom 166 (78%) attended follow-up. Log transformed measures of near stereoacuity were available for 150/166 (90%) at both enrolment and outcome (Table ). There were significant improvements in the observation and non-surgical treatment groups. The surgery group overall also showed a similar level of improvement although this was not statistically significant. Of the 6 overcorrected children aged 4 or older at enrolment, 5 (1 with prism) had some stereoacuity at follow-up (with an abnormal head posture), and while stereoacuity was absent in the other, his/her stereoacuity at enrolment was unknown.
Change in near stereoacuity (log transformed) by treatment group
63/371 (17%) participants had treatment directed at improving visual acuity only (i.e. refractive correction and/or amblyopia treatment) within 2 years from enrolment, including 26 participants who were observed at enrolment and subsequently began treatment within 2 years of follow-up.
288/371 participants (78%) had LogMAR measures of visual acuity in each eye at both enrolment and outcome. There were small but statistically significant improvements in mean acuity in the worse eye for all groups apart from the surgical group, presumably reflecting maturational changes in the observation group (Table ).
Change in visual acuity (worse eye) between enrolment and outcome, by treatment group