Studies

Five trials involving cetirizine were reported before 2000⁵⁵ (see Appendix 3).

One new trial by Diepgen and colleagues²⁶⁰ compared cetirizine oral solution (10 mg/ml, 0.25 mg/kg) against placebo twice daily for 18 months. The trial included 817 children aged 1–2 years who had had active eczema for at least a month before recruitment and who had at least one parent who had a history of asthma, atopic eczema or allergic rhinitis. The trial was primarily looking at the rate of development of asthma in the infants but also assessed the effect of the intervention on the eczema.

One new trial by Capella and colleagues²⁵¹ compared oral montelukast against a ‘standard’ treatment combination of clarithromycin, cetirizine and mometasone furoate. No data were presented for cetirizine treatment alone and so it is impossible to draw any conclusions about the use of cetirizine for eczema on its own from this trial.

Assessment of risk of bias

Table 54 provides the risk-of-bias assessment for the new study.

TABLE 54

Cetirizine: risk of bias of the included study

Benefits

Diepgen and colleagues²⁶⁰ did not provide many methodological details in the trial report, instead referring the reader to a previous publication containing this information.²⁶¹ Cetirizine was described as having a similar appearance and taste to the placebo. The severity of eczema and use of mild and moderate to potent corticosteroids (secondary outcomes) did not significantly differ between the cetirizine group and the placebo group during the trial. The only significant differences between the groups were a lower use of other oral H1 antihistamines in the cetirizine group and a lower rate of development of urticaria in the cetirizine group, with the latter not being listed as an outcome of the trial in the report.

Harms

Very few details were provided about the adverse events recorded in the trial by Diepgen and colleagues²⁶⁰ apart from the levels of urticaria, for which there was a beneficial effect of treatment with cetirizine.

Overall implications for research and practice

The large, long-term trial by Diepgen and colleagues²⁶⁰ failed to find any significant difference between the groups in the severity of eczema or any topical corticosteroid-sparing effect from the use of 0.25 mg/kg of cetirizine twice daily. The lower rate of use of other H1 antihistamines in the cetirizine group makes it difficult to draw conclusions as antihistamines can be used to treat a variety of other, mostly allergic diseases, the level of which may have differed between the two groups, although the levels of sensitisation to milk, egg, grass pollen and house dust mite were similar between the groups.

Studies

Three trials involving loratidine were reported before 2000⁵⁵ (see Appendix 3).

Only one new trial²⁶² used loratidine for the treatment of eczema and this was as the comparator to test the addition of modified Jiawei Danggui Decection (a type of Chinese medicine). As loratidine was used in both groups, this trial cannot be used to assess the effectiveness of loratidine.

Studies

No studies of fexofenadine were reported before 2000.

Two trials involving fexofenadine were reported after 2000. One new trial, by Kawashima and colleagues,²⁶³ compared fexofenadine hydrochloride (60 mg) given twice daily (morning and evening) for 1 week against placebo. The study population of 411 adults had a diagnosis of eczema according to the Japanese Dermatological Association criteria²⁶ and a pruritus score between 4 and 8 after 3 days of placebo treatment prior to enrolment. All participants received placebo for 1 week prior to the trial and used hydrocortisone butyrate (0.1%) twice a day for the placebo period before randomisation and during the treatment period.

A multicentre trial by Nakagawa and Kawashima²⁶⁴ compared fexofenadine hydrochloride (30 mg or 60 mg twice a day, depending on the age of the participant) against another antihistamine, ketotifen (1 mg twice a day), for 4 weeks. In total, 190 children aged 7–15 years with an average score of ≥ 2 for itching in the 3 days before allocation and requiring 0.1% hydrocortisone butyrate on ≥ 70% of their body were randomised.

Assessment of risk of bias

Table 55 provides the risk-of-bias assessment for the new studies.

TABLE 55

Fexofenadine: risk of bias of the included studies

Benefits

The trial by Kawashima and colleagues²⁶³ found a significant benefit of fexofenadine for itching as judged by the participants and also as measured by the ratio of area of pruritus to body surface area, assessed by an investigator.

In the trial by Nakagawa and colleagues²⁶⁴ there were no significant differences between the treatment groups in the mean change in itching score, daily change in itchiness, improvement of rash, participant assessment of the eczema and rate of adverse events.

Harms

In the trial by Kawashima and colleagues²⁶³ the number of participants who experienced adverse events was approximately the same in each group (48 in the fexofenadine group and 45 in the placebo group), with no serious adverse events and one withdrawal because of adverse events in each group. The most common adverse events reported were drowsiness, increases in serum bilirubin and glutamic-pyruvic transaminase, and positive urinary protein. The incidence of these events was similar between the two groups.

There were no serious adverse events reported in the trial by Nakagawa and Kawashima.²⁶⁴

Overall implications for research and practice

Without an objective measure of severity in the trial by Kawashima and colleagues²⁶³ and because of the likelihood that judging the surface area of pruritus could be extremely difficult for an investigator, it is difficult to assess the significance of the efficacy results from this trial. Longer-term trials with at least one validated objective measure of eczema severity are needed. The trial by Nakagawa and Kawashima²⁶⁴ does not provide any evidence of benefit for fexofenadine compared with ketotifen.

There is not yet any positive RCT evidence for this treatment in children, in whom drowsiness is sometimes less of a problem or even a potential benefit for reducing sleep loss as a result of eczema.

Studies

Two trials on ketotifen fumarate were reported before 2000⁵⁵ (see Appendix 3). There were no studies on epinastine hydrochloride reported before 2000.

Two trials involving ketotifen fumarate, one of which used epinastine hydrochloride as the comparator, have been published since 2000. The trial by Nakagawa and Kawashima²⁶⁴ compared fexofenadine against ketotifen fumarate and is discussed in the previous section.

The trial by the Epinastine Hydrochloride Dry Syrup Clinical Study Group²⁶⁵ compared epinastine hydrochloride (10 mg in 1 g of dry syrup) 1.0 g/day (body weight 14–24 kg) or 2.0 g/day (body weight > 24 kg) against ketotifen fumarate (1.38 mg in 1 g of dry syrup) 1.2 g/day (body weight 14–24 kg) or 2.0 g/day (body weight > 24 kg). The treatments were given as dry syrup for 4 weeks. The trial included 163 children aged up to 15 years. The trial was blinded using a double dummy design so that each group received one dose of treatment and one dose of placebo per day. Children who had been using mild or moderate topical corticosteroid for at least 1 week and who had an itching score of ≥ 2 were included.

Assessment of risk of bias

Table 56 provides the risk-of-bias assessment for the new study.

TABLE 56

Ketotifen fumarate and epinastine hydrochloride: risk of bias of the included study

Benefits

This trial carried out in Japan measured itching using three different outcomes and severity using two separate outcomes. The objective of the trial was to prove the non-inferiority of epinastine hydrochloride to ketotifen fumarate. The trial found no significant differences between the treatments at 2 or 4 weeks in terms of itching or severity of rash in a per-protocol population of 148 children.

Harms

There was a high rate of adverse events in the ketotifen fumarate group, with 22 out of 78 participants undergoing events that were considered to be related to the study treatment, compared with 9 out of 84 in the epinastine hydrochloride group. In particular, drowsiness (seven in the epinastine hydrochloride group and 18 in the ketotifen fumarate group) and nasopharyngitis (14 in the epinastine hydrochloride group and 11 in the ketotifen fumarate group) were common problems.

Overall implications for research and practice

The trial by the Epinastine Hydrochloride Dry Syrup Clinical Study Group²⁶⁵ was the only RCT found that investigated the use of epinastine hydrochloride for the treatment of eczema. This trial was short term and did not compare the treatments against a placebo or another non-antihistamine comparator. Both of the treatments considered resulted in fairly high levels of adverse events, most commonly drowsiness and nasopharyngitis. The evidence for a reduction in itching in adults using ketotifen fumarate has previously (published before 2000) been contradictory, with one trial providing evidence of a reduction from baseline after 3 months in adults and another not providing any evidence of a difference in itching or erythema compared with placebo after 4 months in children.⁵⁵ Longer-term studies comparing ketotifen fumarate and epinastine hydrochloride with other active eczema treatments in common use are needed to clarify whether their use for treating eczema is beneficial, especially when weighed against the level of adverse events.

Studies

One ‘double dummy’ multicentre trial conducted in Japan²⁶⁶ randomised 305 children aged 7–16 years with eczema according to the Japanese Dermatological Association criteria²⁶ to olopatadine hydrochloride (10 mg/day) as a tablet or ketotifen fumarate dry syrup (2 mg per day) for 2 weeks. The children had to have an eczema lesion on the head, neck or face that was expected to be cleared by hydrocortisone butyrate ointment and at least a mild pruritus score (score of 2), with the score being different between night-time and daytime. All participants used hydrocortisone ointment twice a day in an observation period before starting the trial.

Assessment of risk of bias

Table 57 provides the risk-of-bias assessment for the new study.

TABLE 57

Olopatadine hydrochloride: risk of bias of the included study

Benefits

The primary outcome of change in pruritus score and the secondary outcomes of change in global score assessed by a clinician and response score for itching, measured by participants on a 5-point narrative scale, were not significantly different between treatments.

Harms

In the olopatadine hydrochloride group, 29 out of 152 (19.1%) participants reported adverse events, including 18 adverse drug reactions. In the ketotifen fumarate group, 37 out of 153 (24.2%) participants reported adverse events, including 10 adverse drug reactions. None of the events were serious or severe.

Overall implications for research and practice

This trial of olopatadine compared with ketotifen fumarate, which was reported to be a non-inferiority trial, did not provide any evidence of a difference between the two treatments in the reduction of pruritus or eczema severity in children. It is impossible to assess whether or not olopatadine has any clinically relevant benefit for eczema compared with other treatments such as topical corticosteroids and emollients. Both treatment groups changed by an average of < 1 point on a 5-point scale for all outcomes measured. This raises doubts about the clinical relevance of both treatments. The lack of change could be because the severity of the eczema was potentially quite low at baseline because of the 1-week treatment period with topical corticosteroids before starting the study treatment.

Studies

One trial involving chlorpheniramine was reported before 2000⁵⁵ (see Appendix 3).

One new trial by Munday and colleagues²⁶⁷ compared chlorpheniramine elixir 2.5 ml (those aged 1–5 years) or 5 ml (those aged 6–12 years) before bedtime every evening with placebo for a period of 1 month. Participants were allowed a second dose after 3 hours. After 2 weeks, if sleeplessness was still present the dose could be doubled to 5 ml (those aged 1–5 years) or 10 ml (those aged 6–12 years) before bedtime every evening. The trial included 155 children aged 1–12 years with atopic eczema. All participants were given Unguentum M emollient (100 g) (Almirall Hermal GmbH), Efcortelan^® (30 g) (GlaxoSmithKline) and hydrocortisone (1%) cream to use as necessary.

Assessment of risk of bias

Table 58 provides the risk-of-bias assessment for the new study.

TABLE 58

Chlorpheniramine: risk of bias of the included study

Benefits

This trial assessed an intention-to-treat population of 151 participants and did not find any significant differences in the investigator- and participant-assessed severity of itching score, assessed on a 5-point scale; participant-assessed sleeplessness because of itching and scratching; participant-assessed daytime drowsiness; investigator-assessed severity of eczema for excoriation; dryness; lichenification; exudation and crusting assessed on a VAS; and adherence with treatment by weighing the medication. The only significant finding was a reduction in erythema in the chlorpheniramine group.

Harms

Twenty participants out of 151 reported a total of 29 adverse events, of which none was serious. The rate of events was the same in the chlorpheniramine and placebo groups. The trial report did not provide any details about the nature of the adverse events. Three participants were reported to have withdrawn because of adverse events.

Overall implications for research and practice

The weight of evidence from one small study published before 2000 and one large study published in 2002 suggests that there is no beneficial effect of chlorpheniramine for eczema, although the studies were both short term. The fact that the large trial by Munday and colleagues²⁶⁷ attempted to detect a beneficial effect using several different outcomes, with none of these showing a beneficial effect, is all the more convincing. As yet, there is no evidence that using chlorpheniramine has a beneficial effect and some fairly strong evidence to suggest that it has no beneficial effect for eczema.

Studies

Four trials involving topical doxepin for eczema were reported before 2000⁵⁵ (see Appendix 3).

One new trial, by Lee and colleagues,²⁶⁸ compared topical doxepin (5%) cream applied four times a day for 7 days against placebo. The trial included 44 adults with eczema who had moderate to severe daily pruritus for at least 1 week before entering the trial.

Assessment of risk of bias

Table 59 provides the risk-of-bias assessment for the new study.

TABLE 59

Doxepin: risk of bias of the included study

Benefits

This small trial by Lee and colleagues,²⁶⁸ carried out in a Korean population, showed a significant effect of doxepin for the relief of itching. The study concentrated on itching outcomes, using two different measures, and also used EASI scores to assess the severity of eczema. As in the previous trials on this treatment,⁵⁵ a significant improvement in pruritus was found but no significant effect on the severity of eczema. A 15.5% improvement in pruritus on day 1 in the doxepin group was found, rising to 42.6% by day 7. This was reported to be statistically significant in favour of doxepin but the data for the placebo group were not reported.

Harms

The most common adverse event was erythema and xerosis at the site of application, which affected five participants in the doxepin group and three participants in the placebo group. Drowsiness was also a problem for two participants in the doxepin group, with one of these participants withdrawing from the trial.

Overall implications for practice and research

The results of this trial in a Korean population are in agreement with those of the previously reported trials by Drake and colleagues,^269,270 which had a very similar protocol. This trial also looked at 7 days of treatment and noted problems with drowsiness and application site erythema and xerosis. These additional data expand the evidence base for this treatment by confirming the results of the trials by Drake and colleagues.^269,270 Future trials could focus on the long-term relief of itching and levels of adverse events in comparison with other active treatments for the relief of itching, although it seems unlikely that these will be pursued. A clearer picture in specific groups such as infants and young children would be valuable to inform the best use of this treatment.

Studies

Ten trials of sodium chromoglycate for eczema were reported before 2000⁵⁵ (see Appendix 3).

One new trial, by Stainer and colleagues,²⁷¹ compared topical sodium chromoglycate (4%) (Altoderm^™; Thornton & Ross Ltd) against topical lotion vehicle, applied twice daily for 12 weeks. The trial included 114 children aged 2–12 years who were diagnosed with atopic eczema according to the UK Working Party’s criteria⁹ and who also had a SCORAD score between 25 and 60. The same diagnostic criteria were required at both of two assessments carried out 14 days apart. Also, overall skin condition and itching were required to be assessed as a score of at least 2 on a scale of 0–3 on at least four separate days within the 14-day baseline period.

Assessment of risk of bias

Table 60 provides the risk-of-bias assessment for the new study.

TABLE 60

Sodium chromoglycate: risk of bias of the included study

Benefits

The trial by Stainer and colleagues²⁷¹ showed a significant difference in the severity of eczema in favour of sodium chromoglycate as measured by SCORAD scores and the participant-assessed overall skin condition scores. Interestingly, the participant-assessed itching and sleep loss scores were not significantly different between the topical sodium chromoglycate group and the vehicle group. The lack of reduction in itching differs from the results found in previous studies using different formulations of sodium chromoglycate.

Harms

The rate of participant-reported adverse events was quite high, at 66 out of 114, but there was no difference between the groups (34 placebo group, 32 sodium chromoglycate group). Eleven participants had an adverse event that was considered to be treatment related, four in the placebo group and seven in the sodium chromoglycate group. No severe adverse events were reported; however, there were more withdrawals assessed as being possibly, probably or highly related to the study treatment in the sodium chromoglycate group, with five in the sodium chromoglycate group and one in the placebo group.

Overall implications for research and practice

The body of evidence for topical sodium chromoglycate treatment is still mixed; however, a medium-sized methodologically robust trial in children did not find any beneficial effect on itching or sleep loss. This could be because of the formulation used or the population included in this study, with participants having to have a moderately high level of itching on at least 4 days out of 14 to be included in the study, as in some of the previous trials. The higher rate of withdrawal in the topical sodium chromoglycate group because of adverse events that were assessed as being possibly, probably or highly related to the study treatment needs further research, although there appears to be a low number of events possibly, probably or highly related to the study treatment overall.