No job name

Clinical and Experimental Ophthalmology 2007; 35: 706–712
Original Article
A prospective randomized treatment study comparing threetreatment options for chalazia: triamcinolone acetonideinjections, incision and curettage and treatment withhot compresses Amynah Goawalla MRCOphth and Vickie Lee FRCOphth
Central Eye Unit, ACAD Centre, Central Middlesex Hospital, London, UK
ABSTRACT
Conclusions: To our knowledge, this is the first prospective
randomized study comparing the three methods of chalazia Background: Three methods of treating chalazia were
treatment. Results suggest that a single triamcinolone compared: intralesional triamcinolone acetonide injections acetonide injection followed by lid massage is almost as (0.2 mL of 10 mg/mL), incision and curettage and advice effective as incision and curettage in the treatment of cha- regarding the application of hot compresses to the affected lazia and with similar patient satisfaction and less pain and Methods: This was a single centre randomized treatment
study. Patients with a chalazion underwent either of the three treatment options. Chalazion resolution, pain, satisfac-tion and inconvenience experienced because of treatmentswere the outcomes assessed via a telephone interview at INTRODUCTION
inflammation. Cosmetically, they can be unsightly; occasion- Results: 136 consecutive patients were enrolled into the
ally, if large enough they can obscure vision1 and rarely, they study. At the 3-week follow up, the resolution rates in the can lead to conjunctivitis or cellulitis.2 Patients are usually triamcinolone acetonide injection and surgical treatment initially advised to apply hot compresses (H/C) to the cyst groups were not significantly different from each other at with a wet flannel to encourage it to spontaneously drain.
84% (47/56) and 87% (39/45), respectively (P < 0.001), but Previous studies have found a 25–50% resolution rate with was significantly lower in the conservative treatment group this conservative treatment.3,4 Failing this, they are treated at 46% (16/35) (P < 0.001). Pain scores were higher in the surgically by incision and curettage (I&C) under a local surgical treatment group compared with the triamcinolone anaesthetic injection. In the late 1970s, treatment with local- injection group (P < 0.003). Inconvenience experienced by ized steroid injections was first described.5 Since then, therehave been a few prospective interventional studies investi- patients was reported as significantly less in the triamcino- gating the efficacy, simplicity and safety of intralesional tri- lone group compared with the conservative and surgical amcinolone acetonide (TA) in the treatment of chalazia6–8 treatment groups (P < 0.001). Patient satisfaction scores (Table 1). Local skin depigmentation has been reported fol- were lower in the conservative treatment group compared lowing transcutaneous injections but this adverse effect with each of the other two groups which in turn, were is minimized through transconjunctival injection.9–11 Ben no different from each other regarding this outcome Simon et al.8 suggested that patients preferred steroid injec- tions to surgery. In addition, accidental globe penetration ᭿ Correspondence: Miss Vickie Lee, Ophthalmology Department, ACAD Centre, Central Middlesex Hospital, London NW10 7NS, UK. Email: Received 4 March 2007; accepted 31 July 2007. 2007 The AuthorsJournal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists A summary of previous studies investigating the efficacy of TA in the treatment of chalazia 89.6% (n = 43)54.2% with one injection25.4% with two injections 80% (n = 155)60% with one injection20% with two injections has been reported as a result of a transcutaneous injection of Inclusion and exclusion criteria for study patients To date, there have been no randomized treatment Male and female patients over 18 years of age of all ethnic groups studies, comparing these three standard chalazia treatments Patient with a palpable chalazion on any eyelid which forms the main objective of our study. The secondary Patients with multiple chalazia but only single one on each lid objective was to compare patients’ preferences of the Normal lid anatomy enabling lid eversion and transconjunctival treatment options, documenting pain experienced during and after each treatment, the amount of inconvenience the patients perceived the treatments caused and their overall Patients with chalazia which have atypical features (a recurring level of satisfaction with the different treatments.
chalazion, abnormal surrounding lid tissue, associated loss oflashes)that may indicate suspicion of malignancy Patients allergic to any of the agents being used in the study (proxymethocaine, chloramphenicol, lignocaine, triamci- This was a single centre randomized prospective interven- tional study. A single investigator (AG) administered all the TA injections and I&C. Ethics committee approval was Patients without a palpable lid chalazion obtained prior to commencement of the study. One hundred Patients with multiple chalazia on one eyelid and thirty-six adult patients were recruited into the study Patients unable to give consent (even with the assistance of a over a period of 8 months from August 2005 to March 2006.
Patients with a concurrent eyelid infection (cellulitis or The final cohort consisted of 83 women and 53 men. The inclusion and exclusion criteria for recruitment are set out inTable 2. Patients were consecutively assigned one of threemethods to treat their chalazia in a block randomizationfashion13 whereby blocks of five or 10 patients would con- for 5 min after each ointment application. Occasionally, some secutively undergo one of the three treatment options. These resistance was felt on penetrating the cyst wall with the were I&C under a local anaesthetic, a 0.2-mL transconjuctival injection but in all cases, the TA was injected intralesionally.
injection of 10 mg/mL of TA into the cyst followed by adviceto massage the lid, or advice regarding H/C to the cyst with Incision and curettage
A 1-mL subcutaneous injection of 2% lignocaine with a25 gauge needle was used to anaesthetize the eyelid and Technique of TA Injection
the conjunctiva was anaesthetized with a drop of 0.5% The conjunctiva was anaesthetized with a drop of 0.5% proxymethocaine. The eyelid was everted and a lid clamp proxymethocaine. The eyelid was everted and a 28 gauge placed. A vertical incision through the tarsal plate into the needle on a 1-mL insulin syringe was used to inject 0.2 mL of meibomian gland was made and a curette inserted into 10 mg/mL TA (Adcortyl, E.R. Squibb & Sons Ltd, Uxbridge, the chalazion to brake up the loculations and drain the Middlesex, UK) transconjunctivally into the chalazion as chalazion. Chloramphenicol ointment was then applied to illustrated in Fig. 1. The eye was not padded after the proce- the treated eye and it was double padded. The patient was dure but the patient was given chloramphenicol ointment to advised to keep the eye padded for at least 2 h and was given apply to the treated eye three times daily for 5 days and was chloramphenicol ointment to apply to the affected eye three instructed to apply gentle digital massage over the chalazion 2007 The AuthorsJournal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists Conservative treatment
followed up. The primary outcome measured was whether ornot the chalazion had completely resolved. Complete reso- The patient was shown how to apply heat a H/C (a clean lution was defined as no longer having a palpable lesion at flannel dipped in hot water) to the affected lid and gently the location of the original chalazion. Patients were asked to massaging the lid to encourage the chalazion to spontane- check and report over the telephone whether or not they ously drain. They were instructed to spend 5–10 min apply- could feel a lump when palpating over the lid. Although less ing H/C to the chalazion three times daily for 5 days reliable than physically following up patients, this method of followed by chloramphenicol ointment which was given to follow up was used because of the high attendance failure rate at Central Eye Unit. Secondary outcomes measuredwere as follows: Assessment of outcomes
The perceived level of pain experienced during and Patients were followed up via a telephone questionnaire (Table 3) 3 weeks after their treatment. The time frame and The perceived level of inconvenience the treatment and mode of follow up was decided upon based on the findings and design of Simon et al.’s study in which the average time to The overall level of satisfaction with the treatment resolution of chalazia after one TA injection was 2.5 weeks.
(taking into account the success of the treatment, the In this study patients’ visual acuities and intraocular pressures pain the treatment caused, the time and inconvenience (IOP) were also followed up, and no significant change in the treatment took in terms of the burden of aftercare IOP of visual acuities was found in patients as they were and interruption to daily activities such as driving).
Telephone questionnaires were used because of high failure rates of patients attending follow-up appointmentspreviously noted in our minor operations clinic. Fortunately,no patients were lost to follow up. We used a simplifiedversion of the 11-point Numerical Rating Scale14 wherebypatients were asked to rate the level of pain they experiencedduring the treatment they underwent for their chalazia andover the hours and days following the treatment. Patientswere asked to score their overall experience of pain on a scaleof 0 to 10 whereby 0 meant that no pain was experiencedwhatsoever and 10 equalled ‘the worst pain they had everexperienced’. For simplicity, a scale of 0 to 5 was used toscore questions two and three of the secondary outcomemeasures whereby 0 was equivalent to ‘no inconveniencecaused’ for question two, or ‘completely unsatisfied with the Figure 1.
Transconjunctival injection of Triamcinolone into the treatment’ for question three. Conversely, a score of 5 signi- fied that the treatment and aftercare incurred ‘excessive Follow-up telephone questionnairePlease circle the answer indicated by the patient over the telephoneQuestions1. Has your eyelid chalazion completely resolved? 2. On a scale of 0–10 how would you describe the pain you experienced during and after your treatment during the postoperative period 3. On a scale of 0–5, how burdensome was your treatment and aftercare in terms of causing you inconvenience? (For example, were you able to drive or return to work straight afterwards? Did you have difficulty applying the ointment and/or massage to the treated eye athome? Did your treatment restrict you in any other way?)(Not burdensome at all) 4. On a scale of 0–5, how would you rate your overall level of satisfaction with the treatment you received? (You may want to consider how successful the treatment was, how restrictive the treatment and aftercare was, how painful it was and whether you had significantbruising after the treatment.)(Completely dissatisfied) Journal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists Figure 2.
Patients were block randomised into one of three treatment Follow up via telephone after 3 weeks to find out if eyelid cyst had completely resolved If cyst had completely resolved then patient If the cyst had not completely resolved then the patient was offered a clinic appointment 3 weeks from the telephone call for a second treatment Follow up via telephone after 3 weeks to find out if eyelid cyst has completely resolved If the cyst had not completely resolved then patients were referred to the oculoplastics clinic for further assessment and treatment from there inconvenience to the patient’ (for question two) and ‘com- treat all their chalazia. The rationale behind this is first, that plete satisfaction with the treatment and aftercare’ (for ques- treating chalazia in close proximity to one another might introduce a confounding factor into the study, influencing Patients who reported that their chalazion had com- resolution outcomes. Second, patients with multiple chalazia pletely resolved after their first treatment were discharged.
suitable for inclusion in the study had to receive the same All the remaining patients (including those patients whose treatment for all their chalazia, otherwise it would not have chalazia had partially resolved) were offered a 3-week been possible for them to score the secondary outcomes follow-up appointment (6 weeks after their first treatment) regarding inconvenience caused by their treatment and after- for a repeat of their original treatment, apart from patients care and their overall satisfaction with their treatment.
who had been assigned to conservative treatment. These However, in order to enable an independent samples com- patients were retreated by I&C at this second clinic appoint- parison, even if a patient was treated for multiple chalazia, ment and were followed up 3 weeks later via a second tele- he or she was included as one sample in the cohort.
phone call (at 9 weeks). If the chalazion had still not resolved Such patients therefore had to be excluded from the study by this point, these patients were booked into an oculoplas- if the outcomes of the treatments on different eyelids were tics clinic for assessment and further management of the chalazion. (See Fig. 2 for a summary of this treatmentpathway.) Statistical analysis
There were four outcome variables, the primary outcome Multiple chalazia
being whether the chalazia had resolved and three secondary Patients with more than one chalazion on the same eyelid outcomes (detailed as described earlier), namely: the were excluded from the study but patients with more than patient’s pain experience, the inconvenience caused to one chalazion, each on different eyelids, were included in the the patient and the patient’s overall satisfaction with the study and were assigned to one of the three treatments to 2007 The AuthorsJournal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists Overall differences between the three treatment groups I&C, incision and curettage; IQR, interquartile range; TA, triamcinolone acetonide.
Whether the chalazia had resolved was measured as either yes or no, and can be viewed as a categorical variable. The Specific comparisons between pairs of groups chi-squared test was used to compare if there was an overalldifference in the outcomes between the three groups. Sub- sequently, the chi-squared test was also used to examine ifthere was a difference between any two treatment groups.
<0.001
<0.001
These subsequent analyses required further statistical tests, <0.001
<0.001
<0.001
<0.001
and because several comparisons between groups were made <0.001
<0.001
there was more chance of finding a difference between treat-ments due to chance. Therefore, this was accounted Bold figures indicate statistically significant (P < 0.001). I&C, for by performing a Bonferroni correction to the analysis incision and curettage; TA, triamcinolone acetonide.
results. This involved multiplying the P-values resulting fromeach test by the number of comparisons made (three in thiscase).
in the I&C group after a second treatment and therefore none The other three outcomes, patient pain, inconvenience of these patients needed to be referred to the oculoplastics and satisfaction, were measured on an ordinal scale. There- clinic. All patients in the H/C group whose chalazia fore, the Kruskal–Wallis test was used to compare the overall remained unresolved at their telephone follow up (19/35) difference in scores between the three groups. Subsequently, were given follow-up appointments for I&C of their chalazia.
the Mann–Whitney test was used to compare between pairs Seventeen of these 19 patients subsequently underwent suc- of treatments. Again a Bonferroni correction was applied to cessful I&C with complete resolution of their chalazia and allow for the fact that several comparisons are made.
the remaining two were lost to follow up after referral to the The significance of the results was determined from the P-values which result from each analysis. A P-value of <0.05 Chalazion resolution rates and the median as well as inter- was usually regarded as evidence of a statistically significant quartile ranges of the other outcomes measured are summa- rized in Table 4. Specific comparisons between pairs ofgroups were made using the chi-squared test for chalazionresolution and the Mann–Whitney test for all other outcomes. These results are summarized in Table 5. The Chalazion resolution
figures presented the Bonferroni-adjusted P-values for eachpairwise comparison.
The results showed a significant difference in chalazion reso-lution between H/C group and each of the other two groups,but no significant difference between the TA and I&C groups.
After a single treatment session, the TA and I&C groups Pain experienced during and after treatment
performed the best with 84% (47/56) and 87% (39/45) ofchalazia resolved in each group, respectively, compared with There was a significant difference in pain between all pairs only 46% (16/35) in the H/C group. Of the remaining 16% of treatment groups. The pain scores were highest in the (9/56) of patients whose first TA injection was unsuccessful, I&C group with a median score of 7, followed by a median a further three chalazia resolved after a second TA injection, score of 6 in the TA group and lowest in the H/C group increasing the overall resolution rate with TA to 89% (50/56) with a median score of 0. There was a significant difference in this treatment group. The remaining six patients in the TA in pain scores reported by the I&C group compared group were then treated successfully by I&C in the oculo- with the TA group which was statistically significant plastics clinic. The resolution rate increased to 100% (45/45) Journal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists Patient inconvenience
patients about inconvenience of treatment, the experiencethroughout the procedure and the subsequent recovery, The results for inconvenience indicated that there was a Numerical Rating Scores were used to measure these out- significant difference between the TA group and each of the comes as this method was found to be superior in a previ- other two groups. The TA group had the lowest inconve- ous study comparing it with the Visual Analogue Scale and nience scores, with a median value of 2 in this group. There the Verbal Rating Scale.14 The study deemed the Numerical was no evidence of a significant difference in inconvenience Rating Scale to be superior in measuring outcomes such between the H/C and I&C groups. The median scores for as pain in treatment evaluative trials because of its ease of these two groups were 4 and 3, respectively.
scoring for patients, a numerical system being easy toanalyse and because it was found to yield consistent Patient satisfaction
The transconjunctival route of TA injection was also There was a significant difference in satisfaction between the found to be safe as this route appears to avoid localized skin H/C treatment group and each of the other two groups. The depigmentation8,9 or inadvertent penetration of the globe.
H/C group had a median satisfaction score of 2, which was Ho7 documented that two out of the 48 patients that under- significantly lower than the TA and I&C groups, which both went subcutaneous intralesional triamcinolone injection were affected by localized skin depigmentation whereas inour study, none of the 56 patients that underwent this pro- Adverse events
cedure suffered this adverse effect. Our cohort consisted of None of the patients reported any adverse events such as skin patients with a variety of skin pigmentations, of whom at depigmentation, decreased vision from globe penetration least 50% were Afro Caribbean or Asian. Therefore, this or significant haemorrhage following either I&C or TA study supports the notion that the transconjunctival route of TA injection minimizes the risk of localized skindepigmentation. One possible explanation for this could bethat although the injection is aimed at being deposited DISCUSSION
intralesionally, occasionally a small portion of TA may be The results of this study suggest that a single transconjunc- deposited by the needle on the way in or out of the injection tival 0.2-mL injection of 10 mg/mL of TA followed by repeated digital massage at home, has efficacy comparable Previous studies investigating the efficacy of TA in the with I&C in the treatment of chalazia and with similar patient treatment of chalazia have used varying concentrations of satisfaction and less pain and patient inconvenience.
the drug. Whereas in Ho’s study,7 the concentration of TA Incision and curettage of chalazia is one of the most used was 10 mg/mL, in Simon et al.’s study,8 a dilution of common minor operations performed in ophthalmology 40 mg/mL was used injected. We opted to use the lower departments and is often a ‘one-stop’ procedure on the day of concentration of steroid in our study. The rationale for this is the patient’s first outpatient visit. In practice we found that that by using the minimal therapeutic concentration of TA, many patients were not prepared to undergo an operative we would minimize the risk of localized skin depigmentation procedure the same day, or drove to the eye clinic so could in a cohort which we expected to be multiethnic, reflecting not have their eye padded afterwards and had to rebook their the local population. Indeed, an even lower concentration of appointment for the procedure with implications on resource injected TA (5 mg/mL) has been reported as being efficacious management. Moreover, chalazia tend to occur in a younger in a previous smaller study.16 Further studies investigating age group who may have a substantial psychological aversion how the efficacy of intralesional triamcinolone varies with to surgery (which involves an anaesthetic injection plus the different concentrations are required to answer this impor- surgical procedure) as opposed to an injection only. A Cana- dian survey of ophthalmologists had suggested that cha- The 46% baseline resolution rate in the conservative lazion surgery should not be trivialized and should be treated group was consistent with previous reports of chalazia reso- with the same respect as any other ocular surgery.15 Previous lution without medical or surgical intervention. Resolution studies4–7 had suggested that TA injections, although effec- of chalazia through this management option may indeed tive frequently, needed to be repeated before complete cha- depend on the chronicity of a chalazion before H/C are lazion resolution is achieved. Multiple injections would applied. To date, there are no published surveys investigating affect the convenience of and pain induced by the treatment.
Conservative treatment with H/C is unlikely to induce pain The use of antibiotic ointment containing steroid was not but has a significantly lower resolution rate, and the treat- investigated as one of the treatment options in this study.
ment may impact on the lifestyle of the typical younger Although topical steroid ointment may theoretically pen- busy/employed patient who may have difficulty in comply- etrate the chalazia by infiltrating through the conjunctiva into the tarsal plate, this would also inadvertently signifi- We have attempted to quantify these elements by incor- cantly increase the risk of raised IOP in a cohort of patients porating this into our follow-up questionnaire in asking the whose IOP was not specifically being followed up.17 2007 The AuthorsJournal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists The lower pain and inconvenience levels reported by the 4. Cottrell DG, Bosanquet RC, Fawcett IM. Chalazions: the fre- TA group compared with the I&C group are indicative of its quency of spontaneous resolution. Br Med J (Clin Res Ed) 1983; patient friendliness. The pain scores were taken 3 weeks 287: 1595.
after the intervention in order that the postoperative/post- 5. Pizzarello LD, Jakobiec FA, Hofeldt AJ, Podolsky MM. Silvers.
injection pain was included in this scoring. Unfortunately, Intralesional corticosteroid therapy of chalazia. Am J Ophthalmol this may have meant that the score may have been influenced 1978; 85: 818–21.
6. Watson AP, Austin DJ. Treatment of chalazions with injection by whether the treatment was successful.
steroid suspension. Br J Ophthalmol 1984; 68: 833–5.
Chalazia mimicking malignant lesions such as sebaceous 7. Ho SY, Lai JSM. Subcutaneous steroid injection for chalazia: cell carcinoma have been well described but fortunately are prospective series. HKMJ 2002; 8: 18–20.
extremely rare.18 Although largely unknown, the effect of 8. Ben Simon GJ, Huang L, Nakra T, Schwarcz RM, McCann JD, injecting steroid erroneously into a sebaceous cell carcinoma Goldberg RA. Intralesional Triamcinolone Acetonide Injection could, at worse, mask its presence leading to a delayed or for Primary and Recurrent Chalazia: is it really effective? Oph- even missed diagnosis and, in turn, a disastrous clinical thalmology 2005; 112: 913–17.
outcome! Therefore, it is important to ensure that all chalazia 9. Ahmad S, Baig MA, Khan MA, Khan IU, Janjua TA. Intra- being considered for TA injection have no atypical features.
lesional corticosteroid injection vs surgical treatment of In addition to having a resolution rate comparable with chalazia in pigmented patients. J Coll Physicians Surg Pak 2006; conventional I&C, triamcinolone injections theoretically 16: 42–4.
avoid excessive bruising of the lid as it is far less traumatic.
10. Kligman AM, Willis I. A new formula for depigmenting human Not needing an eye-pad post treatment means that patients skin. Arch Dermatol 1975; 111: 40–8.
can drive and resume their daily activities almost immedi- 11. Cohen BZ, Tripathi RC. Eyelid depigmentation after intrale- ately following treatment. Intralesional triamcinolone injec- sional injection of a fluorinated corticosteroid for chalazion. Am
J Ophthalmol 1979; 88: 269–70.
tions for chalazia have obvious economic and practical 12. Nawrocki J, Swietliczko I, Nawrocka Z. Pars plana vitrectomy advantages for the health-care provider as its cost in time and following accidental injection of depot steroids into the vitre- equipment is a fraction of that for conventional surgical ous body. Klin Monatsbl Augenheilkd 1990; 197: 519–21.
treatment. TA injections are therefore a good first-line treat- 13. Schulz KF, Grimes DA. Generation of allocation sequences in ment option for uncomplicated chalazia and could be admin- randomised trials: chance, not choice. Lancet 2002; 359: 515–19.
istered by trained nurse practitioners in the eye clinic or even 14. Bolton JE, Wilkinson RC. Responsiveness of pain scales: a com- parison of three pain intensity measures in chiropractic
patients. J Manipulative Physiol Ther 1998; 21: 1–7.
15. Smythe D, Hurwtiz JJ, Tayfour F. The management of cha- ACKNOWLEDGEMENT
lazion: a survey of Ontario ophthalmologists. Can J Ophthalmol The authors would like to acknowledge North-west London 1990; 25: 252–5.
Hospitals NHS Trust for acting as sponsors for this study.
16. Mustafa TA, Oriafage IH. Three methods of treatment of cha- lazia in children. Saudi Med J 2001; 22: 968–72.
17. Lee YJ, Park CY, Woo KI. Ocular hypertensive response to REFERENCES
topical dexamethasone ointment in children. Korean J Ophthal-
mol 2006; 20: 166–70.
1. Donaldson MJ, Gole GA. Amblyopia due to inflamed chalazion 18. Shields JA, Demirci H, Marr BP, Eagle RC Jr, Schields CL.
in a 13-month old infant. Clin Experiment Ophthalmol 2005; 33:
Sebaceous carcinoma of the eyelids: personal experience with 60 cases. Ophthalmology 2004; 111: 2151–7.
2. Rumelt S, Rubin PA. Potential sources for orbital cellulitis. Int Ophthalmol Clin 1996; 36: 207–21.
3. Perry HD, Serniuk RA. Conservative treatment of chalazia.
Ophthalmology 1980; 87: 218–21.
Journal compilation 2007 Royal Australian and New Zealand College of Ophthalmologists

Source: http://www.eyesthetics.co.uk/images/pdfs/New4-prospective%20randomized.pdf