750
tral retinal capillary shutdown. Surgical sheathotomy tery and the vein at the site of crossing can be as thin as
may be a logical approach to the latter subgroup of pa- 15 µm [17]. Thus, with actual surgical instruments vas-
tients; only a few reports have addressed the subject [11, cular trauma can be very difficult to prevent, even for
13, 14].
skilful vitreoretinal surgeons. Furthermore, all novel
In 1988, Osterloh and Charles [13] first reported the techniques have a learning curve and may, in part, ex-
technique of surgical sheathotomy in a patient with plain the surgical trauma to the affected vein. Converse-
BRVO. They experimented the technique in animals, ly, none of the potential complications of vitrectomy in-
then eye bank and inoperable eyes. The patient was a cluding cataract, retinal tear, and retinal detachment, oc-
54-year-old woman with BRVO and a history of 2 to 3 curred in our patients during the 11 months of follow-up.
weeks of distorted vision; she had a visual acuity im- Whether the enlargement of retinal non-perfusion area in
provement from 20/200 to 20/25 8 months after surgery. our patients was the result of the procedure or of a poor
For technical reasons, sheathotomy was to be performed natural history remains questionable.
preferably on crossings concerning first- or second-order
The literature must be interpreted with extreme cau-
arterioles. Despite this relative success, the authors tion due to its lack of substantial analysis. The assessment
did not describe any further similar attempts. In 1999, of any surgical technique remains very difficult and must
Opremcak and Bruce [11] reported a prospective non- take into consideration the fact that spontaneous improve-
randomised study of 15 patients in whom they performed ment can usually be expected to a certain extent [2, 3, 4,
the arteriovenous crossing sheathotomy. The duration of 7,12]. The BVO Study reported that overall, 50%–60% of
visual symptoms due to BRVO ranged from 1 to 12 patients with BRVO will maintain visual acuity of 20/40
months, with an average of 3.3 months. The average fol- or better after 1 year. In the untreated subgroup of eyes
low-up period was 5 months (range 5–12 months). They with BRVO and intact foveal vascularity and in which
reported a constant post-operative reduction of the isch- macular oedema reduced vision to 20/40–20/200, 37% of
aemic area. Of their patients, 53% showed clinical im- eyes gained two lines of visual acuity and 34% had a
provement of more than two Snellen lines and 20% lost 20/40 or better vision at 3 years follow-up [2]. Finkelstein
two or more Snellen lines. Final post-operative visual [7] reported a series of untreated ischaemic macular oede-
acuity was better than 20/200 in 60% and better than ma (capillary non-perfusion) secondary to BRVO in
20/50 in 27% of patients. They did not observe a correla- which 91% of the 23 included eyes improved in visual
tion between duration of the BRVO, severity of intrareti- acuity after a mean follow-up of 39 months: The median
nal haemorrhage, oedema, or ischemia, and final visual initial visual acuity was 20/80 and the median final visual
acuity. Lessening of the capillary non-perfusion was ob- acuity was 20/30. Furthermore, posterior vitreous detach-
served in all cases but not substantially analysed. In ment and vitrectomy per se may affect the prognosis of
2000, Shah et al. [14] reported the long-term follow-up BRVO by means of delaying neovascularisation and pre-
of five patients (five eyes) with visual acuity of less than venting persistent macular oedema [1, 9, 15,16]. In fact,
20/200 presenting with macular oedema secondary to after the onset of BRVO, the clinical course can be deter-
BRVO. In four of the five eyes, vision improved to mined by the collateral drainage capacity from the area
20/40–20/70 at 12 months and to 20/30–20/70 at 7 years with compromised venous drainage to the adjacent areas
of follow-up. In the remaining eye, visual acuity re- of intact venous drainage. This collateral maturation oc-
mained at finger counting secondary to macular isch- curs over a period of 6 to 24 months [5]. Thus, for eyes
aemia.
with macular oedema, it is suggested that photocoagula-
Our series is dissimilar to these reports which demon- tion therapy be delayed for at least 3 months to permit the
strated that visual acuity can improve after sheathotomy: maximum spontaneous resolution of oedema and intra-
None of our three patients improved his visual acuity. In retinal blood. Conversely, it has been speculated by some
two eyes, the visual acuity remained the same (20/80 and authors [11,13] that early surgical decompression might
20/200 respectively); in the third eye with unremitting theoretically lead to less haemorrhage, ischaemia, and
macular oedema, the visual acuity was 20/200 after 11 oedema, with a shorter recovery time. This issue is debat-
months of follow-up.
able, based on spontaneous recovery discussed above,
In two eyes, we observed a transient vitreous haemor- and in the absence of clear-cut benefit of a potentially ia-
rhage on the 1st post-operative day. Although, arteriove- trogenic procedure. The additional risks of retinal vascu-
nous decompression seemed uneventful in the three lar sheathotomy are indeed numerous: full-thickness reti-
cases and we did not observe any intraoperative bleeding nal tear and detachment, vascular bleeding and vitreous
or vascular dilaceration, a vascular trauma is the most haemorrhage, nerve fibre layer defect and scotoma, post-
likely explanation for these vitreous haemorrhages. Like- operative gliosis and traction, and direct vascular injury
wise, in cases 1 and 2, a complete occlusion of the af- [17].
fected vein was present at the end of follow-up, which
Surgical decompression of BRVO is a technically fea-
could also be the result of a vascular trauma. In fact, one sible procedure. Whether sheathotomy is beneficial in
must keep in mind that the common wall between the ar- the treatment of macular oedema secondary to BRVO or
Fittig
