I. M. M. J. Wakelkamp

Surgical or medical decompression as a first-line treatment of optic neuropathy in Graves’ ophthalmopathy? A randomized controlled trial

Objective  Only a small percentage of Graves’ ophthalmopathy (GO) patients develop optic neuropathy with impending loss of visual acuity. Therapy with methylprednisolone pulses is the treatment of first choice in severe and active GO patients. When the effect is insufficient, patients are usually treated with surgical decompression. We investigated whether surgery could become the first‐line treatment, thus preventing treatment with steroids.

Smoking and disease severity are independent determinants of serum adhesion molecule levels in Graves’ ophthalmopathy

Adhesion molecules play a key role in autoimmune disorders, and serum concentrations of soluble adhesion molecules are increased in Graves’ ophthalmopathy (GO). Whether this is due to the strong association with smoking is unknown. It is also not known if the severity or activity of GO determine the serum levels of adhesion molecules. We measured serum concentrations of sICAM‐1, sVCAM‐1 and sELAM‐1 in 62 euthyroid Graves’ patients with untreated GO, in 62 healthy controls matched for sex, age and smoking habits, and in 26 euthyroid Graves’ patients without GO. GO severity was assessed by the Total Eye Score and the activity by the Clinical Activity Score. Adhesion molecules were measured by highly sensitive ELISAs. GO patients had higher levels than controls (median values in ng/ml with range): sICAM‐1 300 [171–575] versus 244 [119–674], P < 0·001; sVCAM‐1 457 [317–1060] versus 410 [238–562], P < 0·001; and sELAM‐1 61 [19–174] versus 53 [23–118], P = 0·021. Euthyroid Graves’ disease patients without GO had levels similar to controls: sICAM‐1 273 138–453), sVCAM‐1 386 [260–1041] and sELAM‐1 46 [22–118]. Smoking had an independent effect and was associated with higher levels of sICAM‐1 and lower levels of sVCAM‐1 in both GO patients and controls; sELAM‐1 levels were comparable. In the 62 GO patients, sICAM‐1 correlated significantly with severity of eye disease (r = 0·40, P = 0·002). No correlation was found with the duration of GO, the Clinical Activity Score or TBII levels. Multivariate analysis of all 150 subjects showed that the presence of GO and smoking are independent determinants of sICAM‐1 and sVCAM‐1 concentrations. In GO patients, the Total Eye Score was a stronger determinant than smoking. It is concluded that (i) smoking is associated with increased sICAM‐1 and decreased sVCAM‐1 levels; (ii) independent from smoking, euthyroid GO patients have higher levels of sICAM‐1, sVCAM‐1 and sELAM‐1 than patients with euthyroid Graves’ disease or healthy controls; (iii) the major determinant of sICAM‐1 in GO patients is the severity of their eye disease.

Author's response: medical or surgical decompression of optic neuropathy in Graves’ ophthalmopathy: the choice remains unclear

We would like to thank Dr Kapamajian for his interest in our study, which still is the first and only randomized clinical trial comparing steroids with surgical decompression in untreated dysthyroid optic neuropathy (DON) 1 We are, however, of the opinion that his criticisms of our study are unjustified. First, Dr Kapamajian proposes that there are two forms of DON: one the result of mechanical compression of the optic nerve by enlarged extraocular muscles (apical crowding), and another without mechanical compression. The latter so-called noncompressive DON is represented by a single patient in the series of Barrett et al . who had an enlarged optic nerve and sheath complex, which the authors hypothesized could be related to inflammation of the optic nerve in the context of orbital autoimmunity. 2 There are no data to support this speculation. Moreover, even in the absence of direct compression of the optic nerve by enlarged muscles, the raised intraorbital pressure in DON itself may interfere with optic nerve function. Consequently to create more space in the orbit either medically (by steroids that will decrease muscle size) or surgically might be both beneficial in DON. All DON patients described by Barrett et al . (including the one with optic nerve enlargement) had a muscular index of ≥ 50, in contrast to what Dr Kapamajian reports. As all our DON patients had apical crowding, our results truly represent the effects of decompression in DON, which in our opinion is always compressive. Second, Dr Kapamajian compares our study results with a large series of DON patients treated surgically in the Mayo Clinics. 3 Such a comparison, however, is completely out of order in view of the big differences in study design. The Mayo Clinics performed a retrospective follow-up study, observational in nature, in DON patients who had been treated previously with steroids, irradiation, cyclosporine or ocular surgery in 61%. Ours was a prospective follow-up study, randomized in nature, in previously untreated DON patients. Third, with regards to the surgical technique we employed in orbital decompression, we have reported the outcome of the coronal approach in DON patients long ago 4 : the success rate in these series of patients (more comparable to the Mayo Clinics patients) was 76%, in good agreement with the improvement in visual acuity in > 70% of the Mayo Clinics patients. The low success rate of surgical decompression in our present study can thus not be attributed to the coronal approach. This is not to say that we haven’t switched to other approaches: the swinging-eyelid procedure is indeed in full swing in our institution. Fourth, Dr Kapamajian calls our intravenous steroid regimen ‘an apparently random regimen’ under reference to a 2005 study by Kahaly. 5 Again, that is besides the point. Dr Kahaly’s protocol of once weekly i.v. methylprednisolone pulses during 12 weeks was applied mainly to patients with moderately severe Graves’ ophthalmopathy, only 11% had DON. Our protocol for DON patients consisting of 1 g methylprednisolone intravenously on three successive days in weeks 1 and 2 is based on a paper by Kendall-Taylor et al . on DON patients. 6