Staphylococcus aureus (S. aureus) infection is perceived not only as a common secondary complication of atopic dermatitis (AD), but also as a culprit in the worsening of this condition. In addition, the recent development of community acquired methicillin-resistant S. aureus (CA-MRSA) has presented a new challenge to our management of AD, both in treatment of acute infections and maintenance therapy. The investigators would like to perform a randomized investigator-blinded placebo-controlled study of children aged 6 months to 17 years with moderate to severe atopic dermatitis with clinical signs of secondary bacterial infection to study: 1) the prevalence of CA-MRSA in our patient population; 2) the relationship of sensitivity of the S. aureus organism cultured from the infected lesion(s) to clinical response to oral cephalexin therapy and severity of the AD; and 3) whether concurrent treatment of S. aureus infection initially with nasal mupirocin ointment and sodium hypochlorite (bleach) baths can result in long-term S. aureus eradication and clinical stability.

Background Atopic dermatitis (AD) is a chronic condition of pruritus and eczematous lesions that affects 15-20% of children. It commonly presents early in life and is associated with other atopic diseases. Pathogenesis is multifactorial with genetic, immunologic, and environmental components. Generally, decreased production of ceramides by keratinocytes in both normal and affected skin causes disruption in skin barrier function, resulting in increased permeability to environmental irritants and allergens, and transepidermal water loss. There also is an immunologic derangement in the body's response to skin injury, manifested by an increased Th2 response in the acute lesions. In addition, the change in incidence of disease by time and climate suggests that environmental factors are important. Staphylococcus aureus (S. aureus) infection is not only a secondary complication of AD, but also a culprit in exacerbations of AD. Of children with AD, 76-100% are colonized with S. aureus, compared to 2-25% of healthy controls. In addition, S. aureus is the most common cause of infected AD. The qualities of atopic skin may explain this high rate of colonization and infection. 1) Patients with AD have poor expression of the natural antimicrobial peptides (βdefensin2 and cathelicidin) during inflammation, likely increasing the risk of cutaneous bacterial and viral infections. 2) The disrupted lipid layer of atopic skin results in low sphingosine, which normally exerts a potent antimicrobial effect on S. aureus. 3) S. aureus contains adhesins, which readily bind to laminin and fibronectin that are exposed in patients with AD and skin injury. Consistently, the extent of S. aureus colonization correlating with the severity of AD has been shown. This may be explained by 4) S. aureus superantigens (Enterotoxins A & B, TSST-1) in the worsening of AD - via the recruitment of T cells & APCs and the upregulation of cytokines without organism elimination. The Th2 cell pathway is preferentially activated, releasing cytokines (IL-4, IL-5, & IL-13), which induces further inflammation and glucocorticoid insensitivity.

First generation cephalosporins have been the mainstay of treatment for acute impetiginized lesions, but the emergence of community acquired methicillin resistant S. aureus (CA-MRSA) has presented a new challenge. Over the past decade, CA-MRSA rates as high as 74% have been found in some regions of the U.S. MRSA has been detected at equal rates in AD and non-AD patients. Although CA-MRSA has greater antimicrobial susceptibility than hospital acquired MRSA, CA-MRSA is often resistant to typical first-line oral agents, including cephalosporins. However, despite the contrary evidence, patients still clinically improve when prescribed first generation cephalosporins. This observation requires confirmation. The deleterious effects of S. aureus in activating AD and observation of improvement with treatment have led researchers to consider eradication (or suppressed growth) of this organism as an important component of treatment of AD. The anterior nares are the major location for S. aureus colonization. Mupirocin ointment represents the mainstay of therapy for eradicating S. aureus nasal carriage, with low resistance rates against CA-MRSA. Recently, sodium hypochlorite (bleach) has become popular with many US pediatric dermatologists. Bleach has long been used safely and efficaciously as a dental antiseptic. This agent has also been shown in concentrations as low as 0.005% to be effective against S. aureus in wounds and ulcers. We have observed that use of sodium hypochlorite improves the clinical appearance of AD, but no studies have been done to confirm this. Several studies have attempted to show that eradication of S. aureus improves the severity of AD. Most studies using topical mupirocin on affected skin lesions for up to two weeks have shown that when S. aureus decolonization can be achieved, patients show clinical improvement. However, skin is recolonized with S. aureus in subsequent months, resulting in worsening disease and making long term treatment difficult. Many studies do not treat both nasal carriage and skin colonization. Although intranasal mupirocin ointment may be sufficient to eliminate S. aureus in healthy hosts, this organism's affinity for the specific features of atopic skin argues for the concomitant use of direct topical antiseptics, such as dilute bleach, in this particular patient population. In addition, most studies have neglected the possibility that family members are likely to be colonized as well, contributing to recolonization. Recolonization can also be affected by failure to continue to aggressively treat atopic skin with appropriate topical therapy during and following eradication. Recently, a small study was performed on adults with AD, using mupirocin, chlorhexidine wash, cephalexin, and potassium permanganate showed significant clinical improvement in subjects. Finally, it is of interest to determine if the exacerbation of atopic dermatitis with CA-MRSA infection is greater than with methicillin-sensitive S. aureus (MSSA) infection. At least one virulence factor, Panton-Valentine leukocidin, has been found in CA-MRSA that is not found in MSSA. Understanding the relevance of the emergence of CA-MRSA to AD is pivotal in appropriately treating secondary infections in patients with AD. Successful eradication of S. aureus colonization from patients with moderate to severe AD may decrease the future number of secondary infections and improve the overall severity of their disease.

Aims of study - #1: Assess the impact of CA-MRSA on secondary S. aureus infections in children with atopic dermatitis. #2: To study whether eradication of S. aureus with nasal mupirocin ointment and sodium hypochlorite baths in conjunction with appropriate skin care regimen with emollients and topical steroids or calcineurin inhibitors can result in long-term S. aureus eradication and clinical improvement. #3: To study the impact of CA-MRSA on overall severity of AD. We will determine the predictive value of CA-MRSA versus MSSA on the overall severity of AD based on the EASI score.

Study population Patients aged 6 months to 17 years with moderate to severe AD by investigator global assessment (IGA) and clinical signs of bacterial skin infection (weeping, crusting, pustular lesions) are eligible. Exclusion criteria include current or recent use (within past eight weeks) of topical or oral antibiotics and allergy to the cephalosporins or mupirocin. Up to 40 patients recruited will be enrolled.

Study Design This study will be a randomized investigator-blind placebo-controlled study. Prior to intervention, qualitative bacterial cultures and sensitivities of the nares and the most severely infected lesions will be obtained from all subjects. Severity of AD will be scored by both the IGA and EASI scoring systems. Patients will continue daily emollient application and clinician-determined topical anti-inflammatory therapy. All subjects will receive cephalexin TID for two weeks. Twenty subjects and their household members will receive intranasal mupirocin ointment BID for five days. The subjects will receive cephalexin 50mg/kg/day (maximum of 2grams/day) divided TID for two weeks. These subjects will continue the five day regimen of mupirocin once monthly for three months and receive sodium hypochlorite baths twice weekly for three months. The other twenty subjects and their household members will receive placebo intranasal ointment with the same schedule. At four weeks and 12 weeks after initiation of therapy, qualitative bacterial cultures and sensitivities of the nares and the most severely superinfected area will be obtained from all subjects. One potential pitfall is the inability to perform quantitative bacterial cultures, as our laboratory cannot do these and the send-out cost is prohibitive; it is possible that the numbers of staphylococcal organisms will decrease, but be undetectable.

Blinding Mupirocin, sodium hypochlorite, and placebos will be stored in identical tubes and bottles. Patients will be randomly assigned numbers 1-40 upon initial participation of study. Based on their number, they will be assigned treatment A (treatment) or B (placebo). The investigator will be blinded.

Assessment & Bacteriology At all visits, the %BSA affected by AD as estimated from 4 body regions (head/neck, upper limbs, trunk, and lower limbs), and the Physician's Assessment of Individual Signs grading the 6 signs of AD (erythema, edema/induration/papulation, excoriation, oozing/weeping/crusting, scaling, and lichenification) will be determined. The results will be used to calculate the EASI score, a validated composite score that ranges from 0 (clear) to 72 (validated very severe). Additional assessments conducted at each visit will be the IGA and the patient's assessment of itch using a visual analog scale (VAS). Adverse events will be recorded at each visit. Swabs will be plated on blood agar and grown for 48 hours at 37oC. S. aureus will be identified by testing for coagulase activity. Antimicrobial susceptibility will be assessed using the agar disc diffusion method. Sensitivity and resistance to several antimicrobials will be determined.

Treatment & Safety One-half cup of household bleach (sodium hypochlorite 6%) will be placed in a full bathtub of water (40 gallons), diluting sodium hypochlorite to a concentration of 0.005%. For most children, ¼ cup per half tub will be added. Patients will soak in sodium hypochlorite baths twice a week for 3 months. Mupirocin ointment or placebo will be applied to the nares for the subjects and families as detailed above. Cephalexin has proven to be safe and efficacious for all ages. Potential adverse effects include abdominal pain, diarrhea, transient elevation of liver enzymes, and allergic reactions. Inappropriate use of cephalexin has led to resistant bacteriologic strains. Mupirocin ointment has been shown to be safe and effective in neonates and children. Adverse reactions to mupirocin include burning, stinging, pain or itching at the site of application. In less than 1% of patients, nausea, contact dermatitis, and skin tenderness/swelling have been reported. While long-term regimens may be associated with development of resistance, short courses of treatment with mupirocin are associated with remarkably little bacterial resistance. Sodium hypochlorite in low concentrations is a safe and widely used agent. Potential AEs include skin and eye irritation. Our previous experience is that most children with AD tolerate bathwater with dilute sodium hypochlorite without complication.

End points Primary outcome measure- change in EASI score from baseline to various timepoints during the study. Safety end points will include the incidence of all AEs reported by patient or parent/guardian, or observed by the investigator.

Inclusion Criteria:

• 6 months to 17 years of age
• Moderate to severe atopic dermatitis

Exclusion Criteria:

• Use of cephalexin or other antibiotic in last 6 weeks
• Allergy to cephalosporins