Dear Editor

We thank Pushpoth and Sandramouli for their wide-ranging critique of our paper discussing previous isotretinoin use in potential military aviator recruits.[1] We cannot agree with their comment that the finding of 2 of 47 candidates, with clinically abnormal dark adaptation (DA) and 11 with electroretinogram (ERG) abnormalities does not justify further screening of this population. Aircrew recruits are a selected population, to a large degree without any significant history of systemic or ocular abnormalities. In balance of probability this selection makes it more likely that the retinal function of these individuals falls within the age-matched normal population, not less. The candidates had, for the most part, applied for a career in aviation long after taking isotretinoin, though we agree that it would have been desirable to prospectively follow the ERG and DA changes before and after treatment.

Pushpoth's and Sandramouli's assumption that isotreinoin-related retinal toxicity is dose-related cannot be made from our retrospective report; largely for the reasons outlined in their comments on incomplete dosage information. We do know that none of the individuals had a history, or family history of retinal abnormality and that clinical retinal and visual field measurements were normal. Colour vision tested normal in all cases with an Ishihara Pseudoisochromic Plate Test and Lanthony 15 hue colour vision test.

Our reference to Oner et al.[2] was not to compare the papers results but to disagree with the finding that all side-effects of isotretinoin were short-lived. Had they performed electrophysiology they might have found a different result.

In this paper we lend support to the findings of other studies that rod function can be adversely affected by previous isotretinoin use[3,4]. As aviation is a night-vision critical profession; we consider it justified at present to check the night vision of aircrew candidates with a history of isotretinoin by dark adaptometry, from a flight-safety point of view. We agree with Pushpoth and Sandramouli that a prospective study of the retinal effects of isotretinoin would shed more light on the problem as much remains to be learnt about the safety of isoretinoin.

Sincerely yours,

Susan P Mollan, Malcolm Woodcock, Peter Good and Robert AH Scott

References

1. Mollan SP, Woodcock M, Siddiqi R, Huntbach J, Good P, Scott RAH. Does use of isotretinoin rule out a career in flying? Br J Ophthalmol 2006;90:957-959

2. Oner A, Ferahbas A, Karakucuk S, et al. Ocular side-effects associated with systemic isotretinoin. J Toxicol 2004; 23:189-195

3. Brown RD, Grattan CEH. Visual toxicity of synthetic retinoids. Br J Ophthalmol. 1989;73:286-288

4. Weleber RG, Denman ST, Hanifin JM, Cunningham WJ. Abnormal retinal function associated with Isotretinoin therapy for acne. Arch Ophthalmol. 1986;104: 831-837

Dear editors

We read with interest the article 'Randomised controlled trial of topical cyclosporin A in steroid dependent allergic conjunctivitis' (Br J Ophthalmol. 2006 Apr;90(4):461-4.) This article evaluated the use of cyclosporine A in treating steroid dependent severe allergic conjunctivitis, and has the merit of a randomized study design, comparing 0.05% cyclosporine versus placebo. The sample size of the study is also one of the largest (n=40). However, only 0.05% concentration of cyclosporine was chosen in this study. A literature search on studies of VKC patients receiving topical cyclosporine showed conflicting results. Initial case series showed promising results on VKC patients with topical cyclosporine 2% [1,2] . Hingorani et al [3] were one of the first to carry out a randomized controlled trial (RCT) (n=21) on cyclosporine 2% on VKC and found improvement in signs and symptoms. This was echoed by another RCT carried out at the Wilmer Eye Institute by Akpek et al. [4] with 22 AKC patients receiving topical 0.05% cyclosporine eye drops compared to placebo, with a marginal decrease in symptoms and signs in the treatment group (p = 0.048 and p = 0.002).

Previous studies revealed that topical cyclosporine could lead to a reduction of epithelial and stromal Class II MHC cells, T cells and IgA and IgG plasma cells in VKC patients, highlighting an immunomodulating effect on cell-mediated and humoral immune responses. The effect on mast cells and IgE mediated allergic responses are not significant, however [5]. Immunohistochemical studies on conjunctival biopsy post-cyclosporine also demonstrated decreased numbers and activation of T cells, with a reduction in T cells cytokine expression [6]. Since VKC/AKC involves both IgE and non-IgE mechanisms, cyclosporine should at least have some clinical effect on VKC. However, whether the effect is significant clinically is another question.

Treatment benefits were seen in patients with keratoconjunctvitis sicca (KCS) or chronic dry eyes (with cyclosporine 0.05% trade name: Restasis)[7]. It remains to be seen if topical cyclosporine could be useful in chronic allergic eye diseases such as VKC and AKC.

References

1. Bleik JH, Tabbara KF. Topical cyclosporine in vernal keratoconjunctivitis. Ophthalmology 1991 Nov; 98 (11): 1679-84.

2. Kaan G, Ozden O. Therapeutic use of topical cyclosporine. Ann Ophthalmol. 1993 May; 25 (5): 182-6.

3. Hingorani M, Moodaley L, et al. A randomized , placebo controlled trial of topical cyclosporine A in steroid dependent atopic keratoconjunctivitis. Ophthalmology 1998 Sep; 105(9): 1715-20.

4. Akpek EK, Dart JK et al. A randomized trial of topical cyclosporine 0.05% in topical steroid resistant atopic keratoconjunctivitis. Ophthalmology 2004 Mar; 111 (3): 476 ¡V 82.

5. el-Asrar AM, Tabbara KF, et al. An immunohistochemical study of topical cyclosporine in vernal keraotoconjunctivitis. Am J Ophthal 1996 Feb; 121 (2): 156-61.

6. Hingorani M, Calder VL, et al. The immunomodulatory effect of topical cyclosporine A in atopic keratoconjunctivitis. Invest Ophthal Vis Sci. 1999 Feb; 40(2) 392-9.

7. Barber LD, Pflugfelder SC, Tauber J, Foulks GN. Phase III safety evaluation of cyclosporine 0.1% ophthalmic emulsion administered twice daily to dry eye disease patients for up to 3 years. Ophthalmology. 2005 Oct;112(10):1790-4.

Jeffrey CF PONG, Jimmy SM LAI

United Christian Hospital HKSAR

Sir,

With great interest we read the letter by Brannan et al, "The use of magnetic resonance imaging in the diagnosis of suspected giant cell arteritis" (BJO 2004;88:1595). The authors conclude that MRI scanning is unable to distinguish between a normal and an unaffected artery and that there is no potential for the use of MRI scanning without contrast enhancement in the evaluation of patients with giant cell arteritis (GCA).

In the past three years we have conducted research including clinical studies addressing this question. Concordantly to the findings in the above mentioned letter we learned that precontrast MRI did not provide sufficient information for differentiation of affected from unaffected segments of the superficial cranial arteries in GCA. However, we were able to demonstrate significantly contrast enhancement and wall thickness increase in affected segments of the superficial cranial arteries in patients with histologically proven GCA (Figure 1). The development of an MR imaging protocol for the acquisition of high resolution images with an in-plane spatial resolution of 195µm x 260µm was has proven to be a reliable tool for the depiction of these small vessels and their walls in great detail. Since enhancement pattern may vary from patient to patient and decrease under corticosteroid treatment we established a four point ranking scale of mural inflammatory changes in GCA [1]. Using 3 Tesla MRI higher signal to noise ratios rendered better image quality as compared to 1,5 Tela imaging. With a large field of view the entire cranial circumference could be depicted in a single scan allowing to study the cranial involvement pattern [2]. In GCA positive patients we found that several cranial arteries were affected simultaneously and often bilaterally. Inflammation of the occipital arteries with spared temporal arteries was also encountered. Based on such cranial involvement patterns, the segment with the most intense inflammatory changes can be determined and recommended for temporal artery biopsy which is considered to be the diagnostic gold standard. However, temporal artery biopsy may be false negative due to the known segmental distribution of mural inflammation. MRI may be of additional diagnostic value to eventually reduce the number of false negative biopsies, especially in patients with unusual distribution of affected arteries.

Figure 1. Enlargements of T1 weighted contrast enhanced spin echo images revealing the frontal branch of the right superficial temporal artery in two patients: No signs of mural inflammation in a patient without signs of vasculitis (arrow in A). Mural contrast enhancement of the thickened vessel wall indicating mural inflammation in a patient with proven giant cell arteritis (arrow in B).

Furthermore, GCA may also involve inflammation of extracranial arteries such as the thoracic aorta and its major branches. Since the diagnostic value of high resolution imaging of superficial cranial artery wall changes is relatively insensitive with respect to imaging delays after contrast agent administration, scanning may be started several minutes after injection [3]. The time remaining could be used for performing a first pass MR-angiography of the thoracic aorta and its large branch vessels for assessing the thoracic, cervical and cranial vasculature [4, -6]. Even abdominal and calf artery involvement has been reported in GCA and may be demonstrated by MRI scanning within the same investigation [7].

After long term steroid medication mural inflammatory signs decrease significantly and eventually vanish [7- 9]. Therefore, we recommend to perfom MR imaging with initiation of steroid treatment or at least within the first few days in order to increase diagnostic sensitivity.

According to our MRI research findings in GCA we conclude that contrast enhanced high resolution MRI has great potential to assist in the challenging diagnosis of giant cell arteritis. It may be used to determine the best suitable site for temporal artery biopsy and eventually reduce the number of false negative biopsy results. Furthermore, the cranial and extracranial involvement pattern can be assessed noninvasively within one single observer independent examination of less than 40 minutes duration.

References

1. Bley TA, Wieben O, Uhl M, Thiel J, Schmidt D, Langer M. High resolution MRI in giant cell arteritis: vessel wall imaging of the superficial temporal artery. AJR Am J Roentgenol. 2005 Jan;184(1):283-7.

2. Bley TA, Wieben O, Uhl M, Vaith P, Schmidt D, Warnatz K, and Langer M. Assessment of the Cranial Involvement Pattern of Giant Cell Arteritis With 3-Tesla Magnetic Resonance Imaging. Arthritis and Rheumatism 2005 Aug;52(8):2470-7.

3. Bley TA, Markl M, Wieben O. Inflammatory hyperenhancement persists in delayed high-resolution MRI in giant cell arteritis. AJR Am J Roentgenol. 2006 Apr;186(4):1197-8.

4. Bley TA, Wieben O, Uhl M, Miehle N, Langer M, Hennig J, Markl M. Integrated Head-Thoracic Vascular MRI at 3T: Assessment of Cranial, Cervical and Thoracic Involvement of Giant Cell Arteritis. MAGMA. 2005 Sep;18(4):193-200.

5. Bley TA, Uhl M, Venhoff N, Thoden J, Langer M, Markl M. 3-T MRI reveals cranial and thoracic inflammatory changes in giant cell arteritis. Clin Rheumatol. 2006 Apr 25; [Epub ahead of print].

6. Markl M, Uhl M, Wieben O, Ness T, Langer M, Hennig J, Bley TA. High resolution 3T MRI for the assessment of cervical and superficial cranial arteries in giant cell arteritis. J Magn Reson Imaging. 2006 Aug;24(2):423-7.

7. Bley TA, Warnatz K, Wieben O, Uhl M, Scholz C, Vaith P, H Peter H, Langer M. High-resolution MRI in giant cell arteritis with multiple inflammatory stenoses in both calves. Rheumatology (Oxford). 2005 Jul;44(7):954-5.

8. Bley TA, Wieben O, Leupold J, Uhl M. MRI findings in temporal arteritis. Circulation. 2005 Apr 26;111(16):e260.

9. Bley TA, Ness T, Warnatz K, Frydrychowicz A, Uhl M, Hennig J, Langer M, Markl M. Influence of Corticosteroid Treatment on MRI-Findings in Giant Cell Arteritis. Clin Rheumatol. 2006 Oct 5; [Epub ahead of print].

Dear Editor,

We read Mollen et al's (1) article with interest where they have concluded that previous isotretinoin use does not cause a clinically significant reduction in night vision in most people and that the retinal toxic effects of isotretinoin may be measurable by electroretinography (ERG) and dark adaptation (DA). While the authors have successfully highlighted the importance of counselling patients for potential irreversible loss of dark adaptation following isotretinoin use, their report, in our opinion, has failed to substantiate the need for routine screening of potential military and civilian commercial aviators.

In their study, 2 out of 47 patients had both abnormal ERG and DA while 11 others had certain abnormal ERG parameters which may or may not be of practical significance. The interpretation of this finding is debatable in the context of this study being a retrospective analysis, where we cannot assess the electrodiagnostic status of the patients in the study group prior to treatment. Only two patients in the study had abnormalities in both ERG and dark adaptation. Those two patients, X and Y, received treatment for a comparatively shorter period of time (8 and 12 weeks respectively) with respect to others (treatment range of 6 weeks to 6 months). Whether these patients had a higher dose of isotretinoin or they had any predisposing retinal problems are not adequately explained in the report. It would have been informative if the authors had compared the dose effect relations among those patients in whom the dose of treatment was known (8 patients with abnormal ERG and 23 patients with normal ERG in the isotretinoin group).

In this particular study, the authors have mentioned a patient who continued to show signs of retinal toxicity 8 years after cessation of treatment, presumably changes in ERG, but there is no description of this patient either in Table 1 or elsewhere in the article. Further, the authors have compared the persistence of retinal toxicity in this particular patient with the study conducted by Oner et al (2). In their study, Oner et al have looked into the visual acuity, anterior segment changes, intraocular pressure, Schirmer's test, tear film break up time, colour vision and changes in microbial flora. They specifically mention in their article that they did not perform any electrodiagnostic studies in their patients. Perhaps it is inappropriate to compare the two studies which have looked at entirely different aspects of side effects of isotretinoin.

Though Mollen et al have mentioned in their methodology that colour vision was tested in their patients, they have not elaborated on the relevant results in the article.

It is interesting to note that the authors have not justified in their main report their recommendation for routine electrophysiological screening for professions that are night vision critical, except in their abstract. It would be appropriate to conduct a prospective study, to look into the electrodiagnostic and other described ocular changes following the use of isotretinoin, to precisely address their question.

S. Pushpoth, S. Sandramouli

Wolverhampton and Midland Counties Eye Infirmary

Wolverhampton, United Kingdom

References

1.S P Mollan, M Woodcock, R Siddiqi, J Huntbach, P Good and R A H Scott; Does use of isotretinoin rule out a career in flying? Br J Ophthalmology 2006; 90: 957-959.

2.Oner A, Ferahbas A, Karakucuk S, et al. Ocular side-effects associated with systemic isotretinoin. J Toxicol 2004; 23:189-95.