Development and evaluation of evidence-based guidelines on contact lens-related microbial keratitis

Available online at www.sciencedirect.com

Contact Lens & Anterior Eye 31 (2008) 3–12 www.elsevier.com/locate/clae

Development and evaluation of evidence-based guidelines on contact lens-related microbial keratitis Lisa Keay c,d,e,1,*, Fiona Stapleton a,b,c,2 a

Institute for Eye Research, Sydney, Australia Vision Cooperative Research Centre, Sydney, Australia c School of Optometry and Vision Science, University of New South Wales, Sydney, Australia d George Institute for International Health, University of Sydney, Sydney, Australia e The Wilmer Eye Institute, Johns Hopkins University, Baltimore, USA b

Abstract Purpose: To assess the response to clinical guidelines based on recent epidemiological studies of contact lens-related microbial keratitis. Methods: Incidence rates for silicone hydrogel extended wear were summarized from recent studies. Risk factors were listed and two key factors identified: overnight wear and poor hygiene accounting for 43% and 33% of risk, respectively. A pre-guidelines, web-based survey was conducted to assess the need for information and level of knowledge on CL-related microbial keratitis (n = 162). Comparisons to a postguidelines survey (n = 51) measured whether new information was acquired through an educational brochure posted to the membership. Results: In the pre- and post-guidelines surveys, 52% (85/162) vs. 43% (22/51, p = 0.3) indicated they did not have enough information on CL-related microbial keratitis. The importance attached to risk factors was generally reflective of the magnitude of risk and practitioners made appropriate recommendations about low risk modalities. Hygiene practices were rated as highly important in both surveys (>90%) with increased awareness of handwashing ( p = 0.03). Overnight wear of hydrogel lenses was considered highly important in 94% (201/213). However, despite evidence to the contrary, only 53% in the pre-guidelines and 58% in the post-guidelines surveys indicated that silicone hydrogel overnight wear was a highly important risk factor. Conclusions: Research evidence has been disseminated to the BCLA membership and responses gathered via a web survey. There are some areas of disconnect between practitioner opinion, as gauged in this survey, and the evidence from the research, particularly with regards to increased risk with overnight wear. Crown Copyright # 2007 Published by Elsevier Ltd on behalf of British Contact Lens Association. All rights reserved. Keywords: Evidence-based medicine; Clinical guidelines; Epidemiology; Microbial keratitis; Contact lenses; Extended wear; Silicone hydrogel

1. Introduction Contact lenses are a convenient and popular means of vision correction. There are estimated to be 125 million [1] contact lens wearers worldwide and almost 3 million wearers in the United Kingdom. Microbial keratitis is a rare but acute disease, where the corneal tissue is invaded by replicating microbes. Intensive anti-microbial therapy is required to halt the disease process and limit corneal scarring * Corresponding author at: Office 930, 550 Building, 550 North Broadway, Baltimore, MD 21287, United States. Tel.: +1 410 614 5636; fax: +1 410 955 0096. E-mail address: [email protected] (L. Keay). 1 PhD, BOptom, BCLA member. 2 PhD, MCOptom, FAAO, FBCLA.

which may affect vision. Even without permanent loss of vision, microbial keratitis is painful, may require hospitalization for treatment, multiple out-patient visits, inability to wear contact lenses, time off work and substantial costs both in treatment and to the individual [2]. From published incidence rates [3–5], and estimating the number of contact lens wearers in the United Kingdom, there are approximately 1200 new cases of contact lens-related microbial keratitis each year. The safety of a contact lens is often gauged by this complication as there is potential for loss of vision. The Oxford Centre for Evidence-based Medicine defines evidence-based medicine as the ‘conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients’. It is in the interests of best patient care [6] that research evidence should

1367-0484/$ – see front matter. Crown Copyright # 2007 Published by Elsevier Ltd on behalf of British Contact Lens Association. All rights reserved. doi:10.1016/j.clae.2007.10.003

4

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

be considered when recommending types of contact lenses and providing advice to contact lens wearers. Large populations are required to study microbial keratitis in contact lens wearers and the first epidemiological studies date back to the late 1980s. These field-based studies of corneal infections secondary to contact lens wear have provided evidence, which has lead to changes to health policy. In 1989, a surveillance study [4] of 5 states in northern United States found a five-fold increased risk of infection with overnight contact lens use. A case-control study in the US [7] published in 1989 also found increased risk of infection with overnight contact lens use. It was this evidence which lead to the Food and Drug Administration in the United States, reducing their approval of disposable contact lens use from 30 nights continuous wear to 6 nights extended wear in 1990. In the UK, similar findings on the risks of overnight wear [8] led to the Department of Health recommendation of signed consent for extended wear of soft contact lenses. These changes in health policy were also reflected in prescribing behaviour, overnight contact lens use which was more common in the 1980s was largely abandoned in the United Kingdom, prior to use of silicone hydrogel contact lenses [9]. There has been renewed interest in this area with the introduction of silicone hydrogel lenses to the marketplace in 1999. The elimination of the hypoxic complications of overnight wear was established early [10–12]. These new materials were comparatively resistant to in-eye spoliation [13] and were proposed for up to 30 nights of continuous wear. Daily-disposable lenses are also of interest having removed the need for lens storage where microbial contamination might occur. Daily-disposable lenses had not been included in the epidemiological studies of the late 1980s to early 1990s. These lenses now constitute up to 1/3 of the contact lens market in the UK and the impact of daily-disposable lenses on the risk of infection can now be examined [9]. Leaders in the field have speculated on the means to examine the risk of new lens modalities [14,15] and several new studies have been convened. In September 2003, a 12month surveillance study commenced in Australia [16] and New Zealand to determine the incidence of contact lensrelated keratitis for the first time in these markets and to establish the changing risks associated with the way contact lenses are used. These studies received a high level of support from practicing ophthalmologists and optometrists in Australia and New Zealand and collaboration from large hospital clinics to supplement case reporting [17]. The information from these local studies together with casecontrol studies from hospitals in London [18,19] and Manchester [20] and from a 5000 patient prospective evaluation of Focus Night & Day contact lenses in the USA [21] constitute a significant body of information on the contemporary risks in contact lens wear. The number of new investigations and scope of recent studies on contact lens-related microbial keratitis means a review of new evidence is timely. Findings from these studies have been appearing in the scientific literature and at

conferences since 2005. Disseminating research findings to health care providers is an important aspect of public health research. The conclusions drawn from epidemiological research are of limited value unless they are made available to the community. As in other fields, while large amounts of funds are directed at clinical research, less effort is directed to ensuring that research findings are communicated and implemented in clinical practice [22]. The dissemination of research findings directly to practitioners has been uniquely funded by the 2006 BCLA Dallos Award on ‘Development and evaluation of evidence-based clinical guidelines on contact lens-related microbial keratitis’. The purpose of this analysis is to measure any changes in prescribing habits or level of knowledge following exposure to this information.

2. Methods 2.1. Surveys and distribution of guidelines Before preparing clinical practice guidelines, the need for the guidelines was assessed [23]. From a scientific perspective, the risk of microbial keratitis with new options such as silicone hydrogel continuous wear and dailydisposable lenses was unknown and could only be gained through large epidemiological studies. However, we were interested to survey contact lens practitioners to gauge the importance of research findings on the risks of microbial keratitis to clinical practice. To establish the opinions and knowledge of the BCLA membership, a pre-guidelines, web-based survey was conducted. The BCLA Dallos award for ‘development and evaluation of evidence-based guidelines on contact lens-related microbial keratitis’ was announced on 1 July, 2005 in the BCLA newsletter. This newsletter was posted to 1697 members and they were invited to participate in the website survey. An email was sent to 615 conference delegates in July and a BCLA update email on July 25th to all members including the website link. This pre-guidelines survey sought to establish the level of knowledge amongst the BCLA membership and level of need for information on microbial keratitis in contact lens wear. The survey also benchmarked the use of hydrogel and silicone hydrogel lenses for extended wear (Fig. 1). Both surveys collected information on practice location, type of practice (independent, multiple, research, hospital or other) and whether they fitted contact lenses. They were asked about whether they provided educational material for their patients, whether they provided an after hours service, if they required informed consent from their patients and what advice they provided, if any, in case of emergencies related to contact lenses. A series of risk factors were listed and practitioners were asked to rate their importance on a 4-point scale (highly important/limited importance/not a risk factor or not important/uncertain). Respondents were asked whether they

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

recommended, prescribed-if-requested or did not prescribe extended wear for hydrogel and silicone hydrogel soft lens materials separately. In addition, they were asked to nominate a low risk form of vision correction for a patient who had a previous episode of contact lens-related microbial keratitis. There was the choice of daily disposables, disposable daily wear, daily wear silicone hydrogel soft, silicone hydrogel extended wear if desired, rigid gas permeable lenses, laser refractive surgery or spectacles only. Finally, information was sought about the need for clinical guidelines on contact lens-related microbial keratitis. Respondents indicated whether they had sufficient information and if not, what type of information they required and the desired format for this information. The post-guidelines survey contained the same questions but in addition asked whether they had read the clinical guidelines brochure, whether they either discussed the data on incidence rates, risk factors for disease or healthcare options in case of an emergency with patients or used the incidence rates and risk factor information in clinical decision making (never/rarely, occasionally or certain patients/all patients). Finally, open comments on the brochure were sought. The full survey was available on the VisionCRC website via direct link as an interactive page. Data were saved directly to a database and data exported at the end of each survey. Ethics approval for this survey was not obtained and consent was implicit in completion of this voluntary survey. Data were stored in de-identified format where practitioners were only known by an unique identification number. 2.2. Generation of guidelines Data were considered from published reports and conference abstracts of epidemiological studies which included silicone hydrogel lenses and incidence rates are summarized in Table 1 [18–21]. Differences in study design and qualitative issues such as validity of control groups and sample size were considered in summarizing the risk with extended wear silicone hydrogel lenses. Information on the factors which contribute to the severity of disease and clinical presentation, symptoms and typical treatment were also reviewed from this family of studies. 2.2.1. Silicone hydrogel extended wear Even without comparison to other contact lens types, the rate of infection appears elevated in silicone hydrogel extended wear in comparison to historical estimates of the Table 1 Incidence of contact lens-related microbial keratitis in silicone hydrogel extended wear Study

Incidence per 10,000

Stapleton 2005 Preliminary data Schein 2005 Morgan 2005

19.3 (13.9–31.3) 18.2 (8.5–33.1) 19.8 (6.4–58.0)

5

rate of microbial keratitis associated with daily wear of hydrogel contact lenses (3–5 per 10,000) [3–5,24]. Considering that recent data shows that silicone hydrogel materials, per se, have not reduced the risk of microbial keratitis in these epidemiological studies (odds ratio 1.0, 95% CI 0.4–2.3) [18], data were presented for overnight wear as an independent risk factor for microbial keratitis. Preliminary data on 82 cases from the first 10 months of a 2 year case-control study conducted at Moorfields Eye Hospital [18] was used to define the magnitude of risk associated with overnight lens wear using multiple logistic regression. Compared to daily wear, wearing lenses overnight up to 6 per year increased risk by 2.0 (95% CI 1.1– 3.72) and wearing lenses overnight >6 per year increased risk by 5.2 (95% CI 2.3–12.0). The key message was that overnight wear remains a risk factor in all lens materials; 2 times increased risk with occasional overnight use and 5 times increased risk with full time extended wear. The available data were reviewed to identify low risk modalities. The incidence of infection with different modalities is best summarized by the Australian case series as the case series was large enough to detect cases with all types of contact lenses. The rate of microbial keratitis in strict daily wear of disposable lenses (1 week, 2 week or 1 month) is very low at approximately 1 in 10,000 annually [25]. Other comparatively low risk modalities in these studies were daily-disposable contact lenses and rigid gas permeable contact lenses. 2.2.2. Risk factors The elements which influence the likelihood of acquiring a microbial keratitis in contact lens wear were divided into modifiable and non-modifiable risk factors (Table 3). Some of the risk factors identified did not persist in multivariate analysis suggesting either that they were not relevant or that the factor was not present in a large enough proportion of the population to maintain statistical power in the multivariate models. It is argued that these should be retained in recommendations, particularly if the associated odds ratio is high, such as in the increased risk when using lenses while on holidays. There are other reasons to consider these factors. Some factors are present in other complications of contact lens wear. For example poor health, specifically upper respiratory tract infection is a factor in contact lens associated corneal infiltrates [26]. Advising against overnight lens wear when unwell might be conservative but prudent advice. Factors such as over the counter/internet purchase of contact lenses further qualify on the grounds of common sense. The logic being that over-the-counter supply does not include education about lens use and individually tailored advice on safe contact lens use. The many risk factors were listed graphically on the brochures with red colours indicating higher levels of risk. While all factors were included, the two main factors which have the strongest influence on the risk of infection:

6

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

overnight wear and poor hygiene, were emphasised. Overnight wear accounted for 43% of the risk of infection and poor hygiene 33% of risk (PAR%). 2.2.3. The outcome of contact lens-related microbial keratitis Most infections are treated successfully with topical antimicrobial therapy. Permanent vision loss of 2 or more lines, due to corneal scarring or the need for surgical intervention to restore vision, occurred in 14% of cases CL-related microbial keratitis in the Australian study (all lens types). This is similar to the rate of vision loss found in other studies (11–13%) [5,27]. and translates to loss of vision of 60 per million contact lens wearers annually (95% CI, 49–78). There are other measures of disease outcome including costs of treatment, costs to the individual and duration of disease. The primary determinant of disease severity is the type of organism involved, where Gram-negative bacteria, Nocardia species, fungi and Acanthamoeba all increase the severity of disease. In addition delays in receiving appropriate treatment is critical [2]. As little as 12 h delay can increase the costs of treating disease and the duration of symptoms. The importance of education about where to seek emergency eyecare is obvious. 2.3. Field testing and evaluation of guidelines The information above was summarized into a colour, double-sided brochure in the first quarter of 2006 in consultation with a graphic designer. The layout of the brochure was reviewed by the authors, a panel of optometrists in private practice in Australia and a group of optometrists working in clinical research. Feedback from these groups was incorporated into the final version. The clinical guidelines were posted at the end of April 2006 to the membership of the British Contact Lens Association (n = 1697). The post-guidelines survey was delivered in the BCLA update special email on May 8th which is sent to the full membership with email contact details (n = 1350). The brochure was also available to download via a VisionCRC website link on this posting. Preliminary results were presented at the BCLA conference in June 2006. A reminder for final responses was sent on September 4th 2006 as part of the BCLA update and data for the post-guidelines survey finalised by mid-September. The number of copies of the clinical practice guidelines accessed via the internet was determined using WebTrends Analytics 8 (WebTrends1 Portland, Oregon, USA) softwear. Requests for reprints via the BCLA were also noted. 2.3.1. Statistical analysis Comparisons were made between the pre-guidelines and post-guidelines surveys using tests of proportions (Fisher’s Exact or Chi-squared tests). Where there was no difference between the two surveys ( p > 0.8), summary data are presented for the two surveys combined. The risk factor

responses were reduced to a binomial response of highly important vs. limited importance, not a risk factor and uncertain, a Fisher’s exact test was then performed. Statistical analysis was performed using SPSS version 14.0 (SPSS Inc., Chicago, IL, USA) and StatXact-3 for Windows (Cytel Inc., Cambridge, MA, USA).

3. Results 3.1. Pre-guidelines survey There were a higher number of respondents to the preguidelines survey (162 vs. 51 respondents). However, the composition of the groups: membership category, type of practice and country where respondents were located was similar (Table 2). Almost all of the respondents (99%, 160/162) were fitting contact lenses at the time of the pre-guidelines survey. A large number provided educational material to their contact lens patients (93%, 151/162) and a similar proportion (93%, 150/162) provided advice on how to act in case of an emergency related to contact lens use. There were a minority (19/162, 12%) who did not prescribe silicone hydrogel extended wear and these practitioners gave different advice to their patients about emergency care. Similar proportions offered an after hours service (32% in both groups). However, those prescribing extended wear were more likely to recommend local accident and emergency departments in case of an emergency related to contact lens wear than those not prescribing extended wear (45% vs. 23%, p = 0.04).

Table 2 Category of membership, mode and location of practice of survey respondents to the pre-guidelines and post-guidelines surveys Pre-guidelines survey(n = 162)

Post-guidelines survey (n = 51)

p value

Category of membership Disp opt Optometrist Technical Student/other

16% 84% 1% 2%

(26/162) (130/162) (2/162) (4/162)

20% (10/51) 77% (39/51) – 4% (2/51)

0.9

Country UK Europe (notUK) Othera Unknown

72% 19% 7% 3%

(116/162) (30/162) (8/162) (5/162)

78% 8% 6% 8%

(40/51) (4/51) (3/51) (3/51)

0.2

Practice type Hospital Independent Multiple/franchise Locum Res/univ/ind/othb

4% 60% 22% 2% 10%

(6/162) (97/162) (36/162) (4/162) (17/162)

10% 67% 8% 2% 14%

(5/51) (34/51) (4/51) (1/51) (7/51)

0.1

a

Other includes Canada, USA, Australia, New Zealand, Nigeria, Israel. Research/university clinic/industry/other (military and multidisciplinary clinics). Disp opt = Dispensing optician. b

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

7

Fig. 1. Project timelines including the two surveys and the circulation of the brochures.

Those not prescribing extended wear were more likely to ask patients to just remove their contact lenses and present to their practice (37% vs. 11%, p = 0.04). Just over half of the respondents to this initial survey (52%, 85/162) indicated that they did not have enough information on contact lens-related microbial keratitis. The largest need was in educational material for patients (Fig. 2). 3.2. Field testing and evaluation of guidelines While the guidelines had been read by 83% of respondents to the post-guidelines survey (43/51), there were still a large number of practitioners who felt they did not have enough information on contact lens-related microbial keratitis (43%, 22/51) and the areas of interest remained similar (Fig. 2). There were some reductions in the proportion requesting most types of information however these did not reach statistical significance. An identical proportion were active in prescribing extended wear of silicone hydrogels (45/51, 88%). There were slightly fewer recommending silicone hydrogel extended wear in the post-

Fig. 2. Proportion of survey respondents who desired specific types of information on contact lens-related microbial keratitis in the pre-guidelines survey (n = 162) and the post-guidelines survey (n = 51).

guidelines survey but this did not reach statistical significance (13/51, 25% vs. 59/162, 36%, p = 0.2). There was a high level of knowledge about risk factors for contact lens-related microbial keratitis in both surveys. There were twenty risk factors listed and there was a maximum of 8% (13/162) in the pre-guidelines survey and 2% (1/51) of responders in the post-guidelines survey who were uncertain about the importance of a risk factor. However, this learning effect from pre- to post-guidelines surveys was not statistically significant ( p = 0.2). The majority of respondents in both surveys considered factors relating to contact lens hygiene as highly important and there were no differences between surveys (Table 3). The vast majority ranked these considerations: case hygiene (99%), storing lenses in saline (95%), inadequate disinfection (97%), use of tap water (99%) and lack of hand-washing (94%) as highly important. In both surveys hydrogel extended wear was noted as a highly important risk factor for microbial keratitis in the majority of respondents (94%). Fewer practitioners felt that overnight wear of silicone hydrogel materials was a highly important risk factor (54%). There is a disconnection between practitioner responses and the evidence that risk is still increased in overnight wear for silicone hydrogel materials (Table 3). Another area where practitioners placed less importance on a risk factor and the research showed it as important were contact lens wear while on holidays. The importance of systemic disease in the risk for microbial keratitis is not highly ranked in this survey despite findings of increased risk of 2–3. The non-modifiable risk factors such as male gender and young age were rated less important and the research data shows moderately increased risk (1.6–2.6). In both surveys, the question was asked ‘If a patient has a history of a serious complication of contact lens wear (microbial keratitis) and wants to return to contact lens wear, would you recommend as the first choice: disposable soft daily wear (DW), silicone hydrogel DW only, silicone hydrogel extended wear if desired, daily disposable, rigid

8

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

Table 3 The proportion of respondents who rated a risk factor as ‘highly important’ from the pre-guidelines and post-guidelines surveys and odds ratios for analysis of risk factors for microbial keratitis in contact lens wear Risk factors

Pre-guidelines a

Post-guidelinesa

p value

Modifiable Hygiene Case hygiene Saline Disinfection Tap water Hand-washing

160/162 152/162 158/162 160/162 150/162

(99%) (94%) (98%) (99%) (93%)

50/51 49/51 49/51 51/51 51/51

(98%) (96%) (96%) (100%) (100%)

0.6 0.5 0.4 0.6 0.03

3.9 (2.7–5.6)

Mode of use O/N use Low Dk O/N use High Dk O/N use O/N use when sick

111/162 152/162 86/162 130/162

(69%) (94%) (53%) (80%)

40/51 49/51 30/51 44/51

(78%) (96%) (59%) (86%)

0.1 0.5 0.3 0.3

16.9 (11.8–24.2) – – –

Occ O/N: 2 (1.1–3.7)

Other factors Systemic disease OSD Smoking Off label useb Internet purchase Holidays/vacation

94/162 154/162 68/162 98/162 – 58/162

(58%) (95%) (42%) (61%)

(57%) (98%) (45%) (73%)

0.5 0.3 0.4 0.08

(43%)

0.2

2.3 – 1.6 – 2.6 15.1

NS

(36%)

29/51 50/51 23/51 37/51 – 22/51

Water contamination Swimming Swimming no goggles Spa/hot tub Shower with CLs

126/162 134/162 98/162 47/162

(78%) (83%) (61%) (29%)

42/51 47/51 24/51 20/51

(82%) (92%) (47%) (39%)

0.3 0.07 0.06 0.1

1.0 5.5 (1.3–23.5) – 0.7 (0.5–1.5)c

NS

4/51 (8%) 26/51 (51%) – – –

0.5 0.04

2.6 1.6 1.7 2.6 2.4

NS

Non-modifiable Miscellaneous Gender (male) Youth High SES <6 months in EW Season (winter)

15/162 (9%) 58/162 (36%) – – –

Univariate OR (95% CI)

Multi-variate OR (95% CI) & PAR%

3.2 (1.7–5.8) 33%

4.5 (2.9–7.1)

(1.3–4.6)

EW: 5 (2.3–12.0) O/N: 43%

(0.97–2.6) (1.1–6.5) (3.3–68.1)

(1.5–4.4) (1.1–2.4) (1.2–2.4) (0.99–6.7) (1.03–5.7)

Contemporary estimates of risk are listed from univariate analysis of the Australia/New Zealand data set and from the post 2000 literature available at the time of generation of these guidelines [18,19,29]. The multi-variate analysis is using the full data set in the Australia national keratitis studies (n = 286 cases of contact lens-related microbial keratitis) and odds ratios for occasional overnight use and extended wear from multivariate analysis of the preliminary data from the casecontrol study at Moorfields Eye Hospital [18]. O/N = overnight, EW = extended wear, Occ. = occasional, OSD = ocular surface disease, SES = socio-economic status, PAR% = proportion attributable risk percent. a Proportion of respondents who rated the risk factor as ‘highly important’. b Wearing lenses for longer periods than that recommended on the label. c For daily wearers only as extended wear correlated with showering in contact lenses.

gas permeable, laser refractive surgery or spectacles only’. There was no difference in the responses in the first and second surveys ( p = 0.8). The majority recommend dailydisposable lenses in this scenario (78/213, 37%). A small number recommended spectacles only (12/213, 6%). Other recommendations were daily wear of disposable (1 week, 2 week or 1 month) lenses (36/213, 17%), daily wear silicone hydrogel (41/213, 19%) and rigid gas permeable (44/213, 21%). A very small number would suggest silicone hydrogel extended wear if desired (2/213, 1%). There were no respondents in either survey who recommended laser refractive surgery in this scenario. The way information was used is shown in Fig. 3. Practitioners were far more likely to discuss information on risk factors for disease rather than incidence rates (73% vs.

24%, discussed with all patients) reflecting greater utility in risk factor information. While not discussed directly with patients, the incidence of microbial keratitis was often considered in making clinical decisions and 57% considered this information for all patients. Advice for where to seek healthcare was discussed with all patients for the majority of practitioners (88%). Comments about the brochures were sought in a free text field. The majority offered no additional comments (32/51) or discussed their choices in contact lens prescribing (6/51). There was one request for information on changing risk in different countries and climates and one request for information on risk in young children. Three comments were critical of the brochure, criticizing the layout of the brochure (1), terminology (1) and confusing information on

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

Fig. 3. Use of different types of information either for discussions with patients or for use in clinical decision making (n = 51).

incidence rates (1). There were a small number of positive comments (5/51): complementing layout (1), graphical content (1) and usefulness (3). The web-page, where the survey was located, was viewed 291 times during the entire survey period. Via this link, the pdf version of the guidelines could be viewed and downloaded. The web-page had limited circulation outside of the BCLA membership. This page was only accessible via the emailed link and there were no links within the VisionCRC website. It is unlikely that this page could be accessed by searching or browsing the VisionCRC website and was most likely accessed by survey respondents. The link where the brochure was available was advertised in reminder notices and this was downloaded 85 times during the survey period. Emailed requests for reprints of the brochure have been made through the BCLA office from Canada, Sweden, France and the UK and are handled on an ongoing basis via the Institute for Eye Research.

4. Discussion The latest epidemiological research data on contact lensrelated microbial keratitis have shown that the absolute risk of disease remains low and essentially unchanged from previous estimates. Strict daily wear use of disposable hydrogel lenses constitutes the lowest level of risk and these are popular options in contact lens practice. International surveys of practitioners in 2005 show that daily wear soft lenses are prescribed in 80% of cases including silicone hydrogel DW (11%) and daily disposable (15%) [28]. The BCLA has a membership of approximately 1800 and comprises optometrists, dispensing opticians, ophthalmologists, industry personnel, research scientists and students. The BCLA annual clinical conference and exhibition is the world’s largest annual dedicated contact lens event. The membership of the BCLA and the annual conference therefore provide an excellent forum to test these guidelines.

9

The surveys reported here of the BCLA membership have demonstrated a high level of knowledge about the risks for developing microbial keratitis in contact lens wear. However there was some disconnect between the perceived importance of overnight wear as a risk factor for microbial keratitis and the evidence from recent research. The research data summarized here presents a clear indication that the risk of infection persists in overnight wear, regardless of the advances in soft contact lens materials. The risks of overnight wear, together with the importance of adequate hygiene are key messages of this campaign. This data predicts that contact lens wearers can reduce their risk by 43% if they remove their lenses when they sleep and if they maintain good hygiene they can reduce their risk by 33%. Silicone hydrogel lenses have been on the UK market since 1999. Allowing time for these lenses to gain market share and for large studies to be planned, the epidemiological studies did not commence data collection until 2003. Only two studies addressing the important issues of incidence and risks of disease were in press when these guidelines were developed: the Manchester hospital based evaluations of contact lens associated infiltrative events [20,29] and the 5,000 patient prospective evaluation of Focus Night & Day lenses in the US [21]. Data from the Australian surveillance study and Moorfields case-control study were only available in abstracts and conference proceedings [16,18,19]. There were however peripheral publications on the Australian and New Zealand studies [2,30] as well as access to data on file due to the affiliation of the authors with this work. While bias towards the authors own research should be acknowledged, any qualitative statements have been made based on sound epidemiological principles. It is not ideal to produce these guidelines before all final study data are available in peer reviewed journals however we would argue that the guidelines were relevant to large numbers of wearers and the body of evidence was substantial enough to justify this endeavour. Contact lens-related microbial keratitis was the subject of a press conference at the 2005 American Academy of Optometry meeting, and data on new lens types been discussed on the podium since 2004 and hypothesized about since the introduction of these new classes of materials. While it is anticipated that the general conclusions will remain the same, it would be worthwhile to revisit the information when more studies are in the scientific literature. The pre- and post-guidelines surveys did not indicate a significant change in either inclination to prescribe overnight wear or perception of risks of overnight wear in silicone hydrogels. It is possible that the information was familiar to these practitioners but that they felt the increased risk was acceptable in terms of the benefits of continuous wear and elimination of hypoxic complications with silicone hydrogel lenses. It is also possible that the information was not well conveyed or that a differing message has been gathered from another source [20].

10

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

The effectiveness of educational brochures has been criticized elsewhere as a means to communicate research information to clinicians [22]. Bero and colleagues [22] evaluated strategies to communicate research data and favoured strategies with communication by local opinion leaders, local consensus processes and better still interactive educational meetings, outreach visits and audit, feedback, local consensus processes or marketing. This research project involved the professional community via the BCLA. The collaboration with a professional organization, survey of its members and involvement of contact lens practitioners in the development and evaluation of these guidelines contribute to peer ownership of the clinical guidelines. Theoretically this might enhance the likelihood of implementation [31]. Our approach in dissemination of clinical guidelines on contact lens-related microbial keratitis had some components of local consensus but relied largely on distribution of an educational brochure and limited access via the internet, conference presentation and completion of a survey. It will be interesting to monitor attitudes to extended wear in the future as the information from this research field evolves and is discussed at conferences by opinion leaders. Surveys of contact lens fitters indicate that the use of extended wear has begun to plateau at approximately 10% of wearers [28]. It is also interesting to note that the range of silicone hydrogel lenses on the market has doubled but new products are aimed at daily wear. Practitioners in these surveys were very familiar with the importance of long established risk factors such as nonsterile rinsing and inadequate disinfection [3,8,32,33], importance of case replacement and cleaning cases [34] but showed increased awareness of the importance of handwashing before handling contact lenses. While practitioners are well aware of these factors and presumably advise their patients accordingly, these factors continue to influence the risks of infection. Compliance to hygiene standards is certainly still an issue in contact lens practice. Other areas where education of clinicians is indicated are awareness of systemic disease as a risk factor for microbial keratitis. In these surveys only 58% thought this was highly important and a 2 increased risk was reported. Compared to hygiene or overnight wear, comparatively few practitioners considered gender (8%) and age (52%) as highly important factors. This may be due, in part, to the fact that age and gender cannot be modified. They are also less potent factors than the two main factors of overnight wear and poor hygiene so lower importance is perhaps justifiable. These surveys indicated a greater utility of risk factor information in comparison to incidence rate data. It might be easier to communicate the magnitude of increase in risk through a particular behaviour than comparing rates of disease. The brochure used pictorial representation of incidence rates, using figures in a crowd. Despite these attempts to make incidence rates more meaningful, these data were still less likely to be discussed with patients.

This brochure captures a large amount of information including statistical analysis. There was debate during the development stages where some suggested removing all graphs and statistics. Final consensus was that this educated audience should be familiar with the way research evidence is presented so that they might also critically appraise the research information. Therefore the statistical analysis was included in the brochure. With similar sentiments, surveys of British general medical practitioners show reluctance in learning how to appraise the evidence themselves (94%). They prefer to utilize summaries generated by others (72%) and evidence-based guideline protocols (84%) [35]. The support of the BCLA to fund communication of research findings rather than pure research may address similar need amongst contact lens practitioners. With further consultation the structure of these guidelines might find an optimal balance between simplicity and scientific content. While the BCLA was an ideal group to field test these guidelines, a limitation of this evaluation, was the low response rate to the two-phase survey. Other surveys of contact lens practitioners commissioned by Eurolens Research where response is voluntary and where there are no incentives, similar response rates are found; typically <20% [9]. Despite these limitations useful information has been derived from this survey. A similar undertaking in the future might include more structured feedback such as focus groups to increase volume of feedback. A desire for clinically relevant information on disease characteristics, differential diagnosis and outcome of disease was requested in the first survey. Differentiation between benign complications of contact lens wear and microbial keratitis, the only sight threatening complication, is of paramount importance to effective treatment of infectious keratitis. Descriptive information on clinical presentation and treatment processes were included in this brochure. The main unanswered need in the project is educational material for patients. This was outside of the scope of this proposal but is an area for future work.

5. Conclusions A review of research evidence has been drafted into a single page handout for contact lens practitioners summarizing latest research findings for the contact lens practitioners and the response to this information assessed. New information included the continued importance of overnight wear as a risk factor for microbial keratitis even with silicone hydrogel materials. In both surveys, a large number of respondents prescribed silicone hydrogel extended wear (88%). New information on the importance of overnight wear as a risk factor with silicone hydrogels was not reflected in survey responses and prescribing choices. It will be interesting to monitor the impact of this information on contact lens practice as this information is

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

debated, discussed at upcoming continuing education events and becomes more widely accepted. The importance of educating patients about good hygiene practices was also evident. Descriptive information was provided to aid in diagnosis of contact lens-related microbial keratitis, a condition for which timely treatment is critical to a good outcome. Reprints of the brochure may be ordered via the Institute for Eye Research in Sydney Australia (f.stapleton@ier. org.au).

Acknowledgements The authors are grateful to Thomas Naduvilath for statistical advice, Narelle Hine and Ray Fortescue for review of the brochure, Selim Soytemiz and Eric Lo for website design and database management and Carol Woollcott for graphic design. This project was funded by the BCLA 2006 Dallos award and ongoing distribution of the brochure is via by the Institute for Eye Research. The authors would like to acknowledge Vivien Freeman and Alison Ewbank for their kind assistance in distributing the brochures and survey requests to the BCLA membership and finally to BCLA members for supporting this project by completing the surveys.

References [1] Barr J. 2004 annual report. CL Spectrum 2005;January 2005. [2] Keay L, Edwards K, Naduvilath T, Forde K, Stapleton F. Factors affecting the morbidity of contact lens-related microbial keratitis: a population study. Invest Ophthalmol Vis Sci 2006;47:4302–8. [3] Lam DS, Houng E, Fan DS, Lyon D, Seal D, Wong E, Hong Kong Microbial Keratitis Study Group. Incidence and risk factors for microbial keratitis in Hong Kong: comparison with Europe and North America. Eye 2002;16:608–18. [4] Poggio E, Glynn R, Schein OD, Seddon J, Shannon M, Scardino V, et al. The incidence of ulcerative keratitis among users of daily-wear and extended-wear soft contact lenses. N Eng J Med 1989;321:779– 83. [5] Cheng K, Leung S, Hoekman H, Beekhuis W, Mulder P, Geerards A, et al. Incidence of contact-lens-associated microbial keratitis and its related morbidity. Lancet 1999;354:181–5. [6] Guyatt GH, O’Meade MO, Jaeschke RZ, Cook DJ, Haynes RB. Practitioners of evidence based care. BMJ 2000;320:954–5. [7] Schein OD, Glynn R, Poggio E, Seddon J, Kenyon K. Microbial keratitis study group. The relative risk of ulcerative keratitis among users of daily-wear and extended wear soft contact lenses. New Eng J Med 1989;321:773–8. [8] Dart JKG, Stapleton F, Minassian D. Contact lenses and other risk factors in microbial keratitis. Lancet 1991;338:650–3. [9] Morgan P, Efron N. A decade of contact lens prescribing trends in the United Kingdom (1996–2005). CLAE 2006;29:59–68. [10] Covey M, Sweeney DF, Terry R, Sankaridurg PR, Holden BA. Hypoxic effects on the anterior eye of high-Dk soft contact lens wearers are negligible. Optom Vis Sci 2001;78:95–9. [11] Keay L, Sweeney DF, Jalbert I, Skotnitsky C, Holden BA. Microcyst response to high Dk/t silicone hydrogel contact lenses. Optom Vis Sci 2000;77:582–5.

11

[12] Papas E, Vajdic C, Austen R, Holden BA. High-oxygen-transmissibility soft contact lenses do not induce limbal hyperaemia. Curr Eye Res 1997;942–3. [13] Jones L, Senchyna M, Glasier MA, Schickler J, Forbes I, Louie D, et al. Lysozyme and lipid deposition on silicone hydrogel contact lens materials. Eye Contact Lens: Sci Clin Pract 2003;29:192–4. [14] Schein O. Assessing the Safety of the New 30-Night Contact Lenses. Eye Contact Lens: Sci Clin Pract 2003;29:157–9. [15] Stapleton F. Contact Lens-Related Microbial Keratitis: What can epidemiological studies tell us? Eye Contact Lens: Sci Clin Pract 2003;29:85–9. [16] Stapleton F, Edwards K, Keay L, Naduvilath T, Dart JK, Brian G, Sweeney D, Holden BA. The incidence of contact lens related microbial keratitis in Australia. IOVS 2005; 46:ARVO E-Abstract 5025. [17] Keay L, Edwards K, Brian G, Stapleton F. Surveillance of contact lens related microbial keratitis in Australia and New Zealand: multi-source case capture and cost effectiveness. Ophthal Epidemiol 2007;in press. [18] Radford CF, Stapleton F, Minassian D, Dart JK. Risk factors for contact lens related microbial keratitis: interim analysis of casecontrol study. IOVS 2005; 45:ARVO E-Abstract 5026. [19] Radford CF, Stapleton F, Verma S, Minassian DC, Dart JKG. Microbial keratitis among users of silicone hydrogel contact lenses: Interim analysis of a case-control study. IOVS 2006; 46:ARVO E-Abstract 5564. [20] Morgan PB, Efron N, Hill EA, Raynor MK, Whiting MA, Tullo AB. Incidence of keratitis of varying severity among contact lens wearers. Br J Ophthalmol 2005;89:430–6. [21] Schein O, McNally J, Katz J, Chalmers R, Tielsch JM, Alfonso EC, et al. The incidence of microbial keratitis among wearers of a 30-day silicone hydrogel extended wear contact lens. Ophthalmol 2005;112:2172–9. [22] Bero LA, Grilli R, Grimshaw JM, Harvey E, Oxman AD, Thomson MA. Getting research findings into practice: closing the gap between research and practice: an overview of systematic reviews of interventions to promote the implementation of research findings. BMJ 1998;317:465–8. [23] NHMRC. How to review the evidence: systematic identification and review of the scientific literature. Canberra, ACT, Australia: NHMRC Handbook Series on Preparing Clinical Practice Guidelines; 1999. [24] Seal DV, Kirkness CM, Bennett HGB, Peterson M. Keratitis study group. Population-based cohort study of microbial keratitis in Scotland: Incidence and features. CL Ant Eye 1999;22:49–57. [25] Keay L, Radford CF, Dart JK, Edwards KFS. Perspective on 15 years of research: reduced risk of microbial keratitis with frequent-replacement contact lenses. Eye Contact Lens 2007;33:167–8. [26] Sankaridurg PR, Willcox MD, Sharma S, Gopinathan U, Janakiraman D, Hickson S, et al. Haemophilus influenzae adherent to contact lenses associated with production of acute ocular inflammation. J Clin Microbiol 1996;34:2426–31. [27] Nilsson SEG, Montan PG. The annualised incidence of contact lens induced keratitis in Sweden and its relation to lens type and wear schedule: Results of a 3-month prospective study. CLAO J 1994;20:225–30. [28] Morgan P, Efron N, Woods C, Jones D, Grein H, Tranoudis I, et al. International contact lens prescribing 2005. CL Spectrum 2006;January 2006. [29] Morgan P, Efron N, Brennan NA, Hill EA, Raynor MK, Tullo A. Risk factors for the development of corneal infiltrative events associated with contact lens wear. Invest Ophthalmol Vis Sci 2005;46:3136–43. [30] Edwards K, Keay L, Naduvilath T, Snibson G, Taylor HR, Stapleton F. Characteristics of and risk factors for contact lens-related microbial keratitis in a tertiary referral hospital. Eye 2007;in press. [31] NHMRC. A guide to the development, implementation and evaluation of clinical practice guidelines. Endorsed November 1998, Canberra, Commonwealth of Australia http://www.nhmrc.gov.au/publications/ _files/cp30.pdf 1999;Date accessed: May 25, 2007.

12

L. Keay, F. Stapleton / Contact Lens & Anterior Eye 31 (2008) 3–12

[32] Stapleton F, Dart JKG, Minassian D. Risk factors with contact lens related suppurative keratitis. CLAO J 1993;19:204–10. [33] Radford CF, Minassian DC, Dart JKG. Acanthamoeba keratitis in England and Wales: incidence, outcome and risk factors. Br J Ophthalmol 2002;86:536–42.

[34] Radford CF, Minassian D, Dart JKG. Disposable contact lens use as a risk factor for microbial keratitis. Br J Ophthalmol 1998;82:1272–5. [35] McColl A, Smith H, White P, Field J. General practitioner’s perceptions of the route to evidence based medicine: a questionnaire survey. BMJ 1998;316:361–5.