Displaying 1-10 letters out of 448 published
Significant design flaws bias the results in favour of aflibercept
Dr Cho and colleagues present data on a very small cohort of patients with wet AMD that have switched treatment from either bevacizumab or ranibizumab to aflibercept. Of note, this subgroup comprised approximately 8% of the total number of patients switched to aflibercept.
Any retrospective review is likely to be heavily biased by the anticipated 'treatment benefit' of a new therapy particularly if, as in this case, the readers of retinal optical coherence tomography (OCT) scans have the ability to manually correct and alter data that were originally generated by semi-automated methods. In this study, the magnitude of change observed in central foveal thickness was of marginal clinical relevance (7.8% reduction from Baseline) after 1 injection and was further attenuated by 6 months; these results suggest that the retinal OCT scan reader was an important source of bias. This view is further supported by the observation that visual acuity, which may be less liable to investigator related bias, remained unchanged throughout.
Retrospective reviews are of scientific value when conducted in a rigourous and independent manner. Selective reporting of data from this study inevitably undermines any clinical conclusions regarding the relevance of switching patients from anti-VEGF therapies to aflibercept.
Conflict of Interest:
I have consulted for a number of pharmaceutical companies including Novartis, the MAH of ranibizumab in the EU.
The long-term psychosocial impact of correction surgery for adults with strabismus
We read with interest Jackson and Morris's response to our letter.
The author's indicated that it was not possible to conduct a repeated measures ANOVA using SPSS. However, SPSS provides several ways to analyze repeated measures ANOVA through the general linear model command. There are several excellent texts that illustrate how to conduct an ANOVA using a repeated measures design in the SPSS environment.
Second, they posed a question about whether it was reasonable to assume that the data collected 18 months post surgically was specifically related to data collected previously. To answer, yes, any time several measurements are collected over time on the same subject, the data points within each subject are related. Therefore the use of statistical procedures that account for this clustering must be used. The fact that the study was exploratory in nature does not preclude the application of basic statistical principles. On the other hand, the authors correctly noted that they had also analyzed the data using a 2x3 design. This approach is reasonable. Unfortunately, the actual p-values were not provided for the readers in the original article or in their response to our editorial.
Conflict of Interest:
Re:Non-inferiority or superiority?
We would like to thank Dr. Geitzenauer for his comments and concerns regarding our study.
As he has pointed out, in a non-inferiority study there is potential for bias which may narrow down the difference in efficacy. Therefore, conducting a superiority study simultaneously would be difficult, unless sufficient quality is ensured during study planning, conduct as well as data analysis.
However, our study in question was a well-controlled study, with a reasonable non-inferiority margin specified in advance, and conducted in accordance with applicable Guidelines. Therefore we believe our study method is acceptable.
The study plan was sufficiently evaluated as a double-masked comparison study, and the study itself was conducted according to the study protocol and in compliance with GCP (Good Clinical Practice). Further, non-inferiority and superiority were verified for different endpoints, Fluorescein staining score and Rose Bengal staining score, respectively, and the methods were specified beforehand in the protocol. In consideration of multiplicity, significance level was maintained by adopting a closed testing method, by which superiority would be verified only if inferiority has already been verified.
EU regulatory authorities have also accepted the simultaneous data analysis for non-inferiority and superiority if it is a well-controlled study, using previously specified analytical methods 1).
Therefore, we believe that the verification methods used in our study are acceptable.
1) Committee For Proprietary Medicinal Products (CPMP) , Point to consider on switching between superiority and non-inferiority, EMEA 2000; July.
Conflict of Interest:
Non-inferiority or superiority?
Takamura et al. report in a recent edition of the BJO about a "multi- center, randomised, double-masked, parallel study" which leads to both a non-inferiority and a superiority claim of the ophthalmic solution being tested (diquafosol ophthalmic solution).(Reference 1) From a methodological standpoint, the conclusions reached by the authors are not in agreement with current accepted standards in the field of randomised controlled trials. It is methodologically unacceptable to draw conclusions about superiority and non-inferiority at the same time using the same set of data. Already at the design stage a decision has to be made whether a non-inferiority or a superiority trial is to be conducted. Hence, claiming both non- inferiority and superiority at the end of such a trial, as tempting as it may be, violates fundamentals of clinical trials methodology. Non-inferiority trials have specific characteristics which make emphasis on rigorous methods even more important than in superiority trials.(Reference 2) In a superiority trial, the objective is to detect a difference, whereas in a non-inferiority trial, the objective is to fail to detect a difference. These different objectives, among others, have important implications with regard to the analysis and conclusion of a trial. It is flawed to analyse and present data mixing up both approaches. Recommendations for improving the quality of reporting of parallel-group randomised trials, noninferiority and equivalence trials are published and freely accessible on the internet. (References 3 and 4) Adoption of the CONSORT statement in the editorial policy is likely to be beneficial to the high standards of the BJO and would certainly be welcomed by this reader.
Wolfgang Geitzenauer MD MSc FEBO
1 Takamura E, Tsubota K, Watanabe H, Ohashi Y. A randomised, double- masked comparison study of diquafosol versus sodium hyaluronate ophthalmic solutions in dry eye patients. BJO; 96(10):1310-5.
2 Fleming TR. Current issues in non-inferiority trials. Stat Med; 27(3):317-32.
3 Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ;340:c332.
4 Piaggio G, Elbourne DR, Pocock SJ, Evans SJ, Altman DG; CONSORT Group. Reporting of noninferiority and equivalence randomized trials: extension of the CONSORT 2010 statement. JAMA. 2012; 308(24):2594-604.
Conflict of Interest:
Risk of cataract for patients with diabetes mellitus
I read the paper entitled "Risk of selected eye diseases in people admitted to hospital for hypertension or diabetes mellitus: record linkage studies" with interest. It elucidated that diabetes mellitus has a risk of several ocular diseases with significance using two big epidemiological data. However, I have two queries on their outcome by selecting the association between cataract and diabetes mellitus.
First, Goldacre et al. described rate ratios concerning several eye diseases caused by hypertension or diabetes mellitus. For example, risk of cataract in diabetes was high presenting rate ratio (95% confidence interval) of 2.95 (2.75 to 3.16) and 2.30 (2.24 to 2.35) in their two epidemiological datasets (1). Although sex, age, year of admission and patients' area of residence was adjusted by matching procedure, their outcome is a hospital-based case control study with no specification on the type of cataract. Mukesh et al. conducted a follow-up study, and diabetes mellitus and having taken calcium channel blockers for longer than 5 years were independent risk factors for posterior subcapsular cataract (2), presenting hazard ratio (95% confidence interval) of 2.9 (1.7-5.1) and 2.9 (1.2-6.9), respectively. In addition, hazard ratio (95% confidence interval) of age by one year increase for posterior subcapsular cataract was 1.09 (1.07-1.1) (2). Effect of confounding variables on the association between cataract and diabetes mellitus are more understandable in cohort study compared with case-control study as mentioned above.
Second, Goldacre et al. described the study limitation for the lack of information on the clinical state of diabetes mellitus including treatment (1). The information on the clinical state of diabetes mellitus is useful in combination with diabetic retinopathy.
The hospital-based case control study has a merit of satisfactory statistical power, and further study on the association between cataract and diabetes mellitus should be conducted including the above mentioned information.
1. Goldacre MJ, Wotton CJ, Keenan TD. Risk of selected eye diseases in people admitted to hospital for hypertension or diabetes mellitus: record linkage studies. Br J Ophthalmol 2012;96:872-6.
2. Mukesh BN, Le A, Dimitrov PN, et al. Development of cataract and associated risk factors: the Visual Impairment Project. Arch Ophthalmol 2006;124:79-85.
Conflict of Interest:
Improving access to eye care in a remote region in Western Kenya
Using a chart review, Nyamori and colleagues estimated the incidence of retinoblastoma in Kenya to be 1:17,000 live births, similar to global estimates.1 They observed that late presentation was common, often attributed to poor awareness or socioeconomic barriers which hinder access to care. We describe a recent project to improve access to eye care for people living in Western Kenya.
IcFEM Dreamland Mission Hospital in Kimilili achieved hospital status in February 20122. With the support of Sabatia Eye Hospital and the Provincial eye surgeon we then began offering affordable eye surgery once a month. Demand for eye screening was high with people travelling long distances to obtain services. In order to reach as many people as possible, IcFEM engages local stakeholders by setting up community leadership structures called Local Transformation Units2. These helped us to obtain the agreement of local Chiefs, Councillors and Public Health Officers to set up outreach clinics in isolated villages and at local markets. Before each clinic, posters were put up in shops, pharmacies, market places and read at public meetings. A team of two nurses and our resident clinical officer then set up a stall where people could have their visual acuity tested (using a chart which did not require literacy), obtain reading glasses or medication, have simple foreign bodies removed or be booked for surgery at the Mission Hospital. Between January and July 2012, 751 people aged 2 to 100 years were screened: 154 at outreach clinics in the villages, 198 at markets in Kimilili, Kamukuywa and Chwele, and 399 at the hospital. Those being considered for surgery were booked for assessment at the hospital before the monthly operating list conducted by a specialist eye surgeon. Overall 115 patients, mean age 68 years, (15% of those screened) underwent surgery: cataract extraction 106, foreign body removal 5, excision biopsy 2, tarsorrhaphy 1, peritomy 1. In spite of electricity cuts and use of the emergency generator, only one patient had a complication (dislocated lens) requiring further surgery. Following an overnight stay, operated patients were given a talk including use of eye drops and booked for follow up after two weeks. Complex cases, including children requiring a general anaesthetic, could be referred to a specialist hospital.
Nyamori and colleagues recommend increasing public awareness of eye problems and availability of treatment. We hope that the return to remote villages of mainly elderly patients, some of whom were previously blind and are now able to see, help in the fields or care for their grandchildren, will contribute to this.
Rebecca Nightingale BSc, Consultant Physiotherapist Jane Dobbs FRCP, Medical Superintendent Clement Kiprop Dip-CMS, Head Clinical Officer
IcFEM Dreamland Mission Hospital PO Private Bag, Kimilili 50204 Kenya
Correspondence: Rebecca Nightingale email: email@example.com
References 1. Nyamori JM, Kimani K, Njuguna MW, Dimaras H. The incidence and distribution of retinoblastoma in Kenya. Br J Ophthalmol 2012; 96: 141-143
2. IcFEM www.icfem-mission.org
Acknowledgements We thank the staff of Sabatia Eye Hospital for their assistance in setting up the eye department and Dr Simon Daniell for installing the hospital eye equipment. We acknowledge the dedicated team of specialist eye surgeons and theatre sisters who with our colleagues at IcFEM Dreamland Mission Hospital run the eye service. Our thanks also go to Dr Pippa Oakeshott for her helpful advice.
Conflict of Interest:
Funders: ROPE (Relief for Oppressed People Everywhere) www.rope.org.uk. IcFEM ( Interchristian Fellowships' Evangelical Mission) www.icfem-mission.org
Response to `Clinical factors associated with malignancy and HIV status in patients with ocular surface squamous neoplasia at Kililmanjaro Christian Medical Centre, Tanzania'
We read your paper entitled `Clinical factors associated with malignancy and HIV status in patients with ocular surface squamous neoplasia at Kililmanjaro Christian Medical Centre, Tanzania' with much interest. It was an interesting insight into the characteristics of OSSN patients in sub-saharan Africa. However, we had some queries and we hope you can share your thoughts on them.
Firstly, we note that patients were selected based on external appearance of lesions and we feel that this creates a selection bias as patients with clinically less prominent lesions are less likely to be recruited into the study. Your groupings of malignant lesions have included mild to moderate dysplasia, many of which may be clinically less prominent than more dysplastic lesions. This creates a bias where patients selected are more likely to be of a higher dysplastic grade. However, we do agree that selecting a large random group of patients from the general population and testing them histologically may not be feasible.
Our second point is that the p value for independent association with HIV status in the text (p = 0.035) is different from that in the table (p = 0.01) Thirdly, we note that you have concluded that HIV positive patients tend to have a higher malignancy grade based on a regression model. However, we were wondering if there was any explanation for the fact that CD4 counts did not show a similar association. Furthermore, you concluded that positive HIV status was associated with longer lesion duration, larger lesion size, leukoplakia and the presence of feeder vessels. However, this raises a few questions. Firstly, most would agree that a lesion of longer duration is more likely to be of a larger size with corresponding feeder vessels and show leukoplakic changes than one present for a shorter duration. If it possible then that, HIV status was associated with the above risk factors(large size, feeder vessels, leukoplakia) as HIV patients themselves were more likely to have a lesion present for a longer duration of time? Also, it is mentioned that women comprised of 69% of HIV patients. We were wondering if in your experience, women in sub-Saharan Africa may have lesser access to medical care which could have contributed to the longer duration of lesions seen in HIV patients.
Lastly, it is stated that HIV patients are more likely to suffer a recurrence. However, if HIV patients are more likely to have a larger lesion for a longer duration, would this contribute to a higher recurrence rate? Also, if HIV status is a postulated risk factor for the development of dysplasia, it is highly conceivable that there are multiple dysplastic foci. It is then possible that a `recurrent lesion' is in fact the progression of a dysplastic lesion of another focus? As such, were all the recurrences in the same spot of excision?
Thank you for your time and we hope to hear your thoughts on the above queries!
1. Clinical factors associated with malignancy and HIV status in patients with ocular surface squamous neoplasia at Kilimanjaro Christian Medical Centre, Tanzania. Makupa II, Swai B, Makupa WU, White VA, Lewallen S. Br J Ophthalmol. 2012 Apr;96(4):482-4. Epub 2011 Nov PMID: 22075543
Conflict of Interest:
Re:Brimonidine induced uveitis: Extent of the problem?
The authors thank the respondent for the observations and comments. Most of our patients were referred to us by other ophthalmologists. Unfortunately, as tertiary referral practitioners to the wider ophthalmic community (uveitis: JCR, glaucoma: WHM and DD) we have no way of knowing how many patients in our community are on brimonidine or how many cases of brimonidine induced uveitis may be seen by other consultants. We agree with the respondent that brimonidine rechallenge testing adds to the evidence for causality. This has previously been performed, in case reports by Byles  (four patients), Goyal  (one patient), Cates  (one patient) and Becker  (one patient). In all of these cases, uveitis recurred on re-exposure. In our case series, all of the patients had field threatening glaucoma, in one case in an only eye. Given the severity of their alphagan side effects we could not ethically request that any patient voluntarily trial re-exposure to the drug when its implication in uveitis (sometimes hypertensive) is already so well documented. 1. Byles DB, Frith P, Salmon JF. Anterior uveitis as a side effect of topical brimonidine. Am J Ophthalmol. 2000 Sep;130(3):287-91. 2. Goyal R, Ram AR. Brimonidine tartarate 0.2% (Alphagan) associated granulomatous anterior uveitis. Eye (Lond). 2000 Dec;14(Pt 6):908-10. 3. Cates CA, Jeffrey MN. Granulomatous anterior uveitis associated with 0.2% topical brimonidine. Eye (Lond). 2003 Jul;17(5):670-1. 4. Becker HI, Walton RC, Diamant JI, Zegans ME. Anterior uveitis and concurrent allergic conjunctivitis associated with long-term use of topical 0.2% brimonidine tartrate. Arch Ophthalmol. 2004 Jul;122(7):1063- 6.
Conflict of Interest:
Brimonidine induced uveitis: Extent of the problem?
Article "Brimonidine (Alphagan) associated anterior uveitis" by McKnight CM et al1 is informative. Cessation of brimonidine eye drops resulted in improvement of uveitis in five cases. This case series has produced further evidence that brimonidine may be responsible for uveitis/ raised IOP in some cases. However a critic may still argue the two events to be coincidental. Unfortunately we have only a few anti glaucoma drugs that can be used in uveitis. For the sake of the rest of glaucoma patients, a few of these patients can be motivated to be volunteers. Restarting the brimonidine in any of these patients and documenting the reappearance of uveitis would produce stronger evidence. Moreover, their 5 patients1 presented with uveitis after using brimonidine for 13, 17, 6, 12 months and 5 years. Earlier reports also suggested that when brimonidine is used, anterior uveitis can occur after approximately 1 year2/ 2 years3 of treatment. Keeping in view the common use of brimonidine, these case reports reflect a very low incidence of uveitis and that too after use for many months. Had authors stated their total number of patients on brimonidine, we would have gained an idea of the frequency/ prevalence of the problem. References: 1. McKnight CM, Richards JC, Daniels D, Morgan WH. Brimonidine (Alphagan) associated anterior uveitis. Br J Ophthalmol. 2012 Jan 18. [Epub ahead of print] 2. Velasque L, Ducousso F, Pernod L, Vignal R, Deral V. [Anterior uveitis and topical brimonidine: a case report]. J Fr Ophtalmol. 2004 Dec;27(10):1150-2. [Article in French] 3. Nguyen EV, Azar D, Papalkar D, McCluskey P. Brimonidine-induced anterior uveitis and conjunctivitis: clinical and histologic features. J Glaucoma. 2008 Jan-Feb;17(1):40-2.
Conflict of Interest:
Re:Re:Efficacy of Silicone Punctal Plugs in Children
We thank Dr Shoaib for his comments. In the results we clearly state 'The indication for insertion was based on the presence of ocular surface changes and poor tear film meniscus, with a previous unsuc- cessful management by lubrication and topical medication alone. Overall, 18 of the 25 patients had a concurrent systemic disorder (table 1).' Schirmer's test is , we believe , a poor test for dry eye in children. It often results in reflex tearing and is of little clinical value in children. Tear break up time is only of value if we know what the normal value is in CHILDREN. This is in fact the subject of our next manuscript which is in review as I write this. In children under 12 years of age and over 4 years of age , the non-invasive tear break up time using the Tearscope ( Keeler , Windsor ,UK) is over 25 seconds ( unpublished data as yet). So in children we looked at tear meniscus and ocular surface changes such as PEE. We not only relied on subjective improvement but also objective signs of improvement of ocular surface changes. As for the comparison of cases of BKC, we can only comment on our own paper which Dr Shoaib cites(1) . Please note that the majority of the children in the 'Punctal Plug' manuscript had a systemic disorder which led to secondary lid and corneal changes. In the article cited regarding BKC (1) many of the children had neovascularisation of the cornea and lubrication is clearly mentioned but not punctual plugs. We clearly state here in the 'Punctal Plug' article that children who failed lubrication were offered plugs. None of the cohort from the 2007 manuscript were in this manuscript. Finally, it is precisely because children can be so difficult to assess that there has been no previous manuscript, to the best of our knowledge, discussing punctual plugs exclusively in children.
1 Jones SM, Weinstein JM, Cumberland P, Klein N, Nischal KK. Visual Outcome and Corneal Changes in Children with Chronic Blepharokeratoconjunctivitis. Ophthalmology 2007;114:2271-2280
Conflict of Interest:
Register for free content
This recent issue is free to all users to allow everyone the opportunity to see the full scope and typical content of BJO.
View free sample issue >>
Don't forget to sign up for content alerts so you keep up to date with all the articles as they are published.