Cellulitis is defined as a bacterial infection of the skin, usually involving the epidermal and dermal layers, with occasional involvement of the subcutaneous fat and lymphatic tissues. It is commonly characterized as poorly demarcated but localized swelling and pain, most often in the lower limbs [1]. Erysipelas is a subtype of cellulitis where the infection is more superficial in the epidermal layer and is often referred to by its clinical name when it affects the face or legs. The presentation of cellulitis also varies in severity and in bacterial species. Most patients only have mild systemic symptoms such as fevers, with only around 10% of hospitalized patients experiencing severe complications such as sepsis and necrotizing fasciitis [1,2]. However, around 18% of initial antibiotic treatment still fails. The most common causative agents include beta-hemolytic streptococci, such as group A and group G, as well as Staphylococcus aureus. The presence of methicillin-sensitive versus methicillin-resistant species of S aureus is also crucial in directing antibiotic treatment [2]. However, because of the nonsterile site of cellulitis infection with the presence of numerous commensal bacteria, empiric treatments are usually started without specimen collection. Microbiological studies are often only conducted in a research setting, to a limited effect, as the presence of commensal colonies can often confound the culture results. In rare cases, through opportunistic infections, fungal species can be implicated as well [3].

Based on the prominent bacterial species possibly found in common cellulitis infections such as beta-hemolytic streptococci and staphylococci, gram-positive antibiotics covering these organisms have been the empiric drug of choice [3,4]. However, a wide range and different classes of antibiotics are available to fill this demand, and there have been no cohesive guidelines for standardization. Beta-lactam inhibitors such as penicillin or different generations of cephalosporins such as cefazolin, cephalexin, or ceftriaxone have been commonly used. Other classes of antibiotics such as macrolides, doxycycline, and clindamycin have all been used and reported in the literature [3]. Additionally, antibiotics used for the broader coverage of methicillin-resistant S aureus–related cellulitis also vary, from vancomycin to trimethoprim-sulfamethoxazole or clindamycin. These antibiotics all differ in their efficacy and side effect profiles, such as Clostridium difficile risks. This lack of standardization and reliance on physician preference points to the need for evidence-based reviews for the optimal treatment of cellulitis [3-5]. These antibiotic options also differ in their routes of administration, with some available in both oral and intravenous (IV) routes, while other antibiotics can only be administered through one of the options. Because of its potential involvement in deeper layers of skin, topical antibiotic administration is unlikely to show efficacy against the infection. Despite some evidence showing that oral therapies are at least as effective as IV administration, some guidelines still recommend the IV routes for high-risk patients, such as those with obesity or venous stasis [3,6].

It is for these reasons mentioned above that we have decided to perform this systematic review. The primary objective of our review is to compare the efficacy of oral versus IV antibiotic administration for cellulitis infections, thereby providing clinicians with evidence-based guidelines for treatment.

We will share the results of this study through peer-reviewed academic publications and conference presentations. We will continue to perform updating the searches in all databases every 2 months for any new data to be included in the systematic review. This project received no funding or financial assistance. Data analysis is currently underway, and the results are expected to be submitted for publication in late November 2023.

This protocol outlines the planned systematic review and meta-analysis to compare the efficacy of oral and IV antibiotic administration in the treatment of cellulitis. Because of the vast selection of options available to physicians and the lack of quality evidence or rigorous reviews, there has been significant variation between guidelines and physician preferences. Some antibiotics may be available for both IV and oral administration, while others may only be suitable for one of the routes. Therefore, we also hope to aggregate the evidence from the literature first and then elucidate the indications for the usage of either route of administration, potentially also comparing their usage for similar infection profiles to assess their efficacy.

Our review will have several strengths and weaknesses. In terms of strengths, we will not be placing any language or time restrictions in our search strategy to ensure broad inclusion of all currently available evidence. We will also include both prospective and retrospective cohort studies in addition to RCTs to increase our pooled sample size. As a weakness, the quality of our meta-analysis may be limited by our inclusion of cohort studies, which is inherently weaker on the hierarchy of evidence with a potentially higher RoB. However, to mitigate this, we will analyze RCTs and cohort studies in separate pools, as well as using tools such as RoB 2.0 and ROBINS-I to rigorously assess the quality of the included studies based on the Cochrane Handbook. We will also conduct a subgroup analysis with only high-quality evidence and compare those with the pooled results. Additionally, because of the variations that exist within guidelines and physician practices, we also expect potentially significant heterogeneity between our included studies. We will assess this heterogeneity quantitatively and report on it in our final review. An additional limitation of our review is that, despite placing no language restrictions, the databases searched are primarily Western, English-based databases, meaning additional non-English data may be missed. We are only focusing on adults aged 18 years or older; hence, the value of this study for patients aged 18 years or younger may be limited.

There have been several attempts at assessing the different routes of antibiotic administration for the treatment of cellulitis. Brindle et al [3] conducted a systematic review on cellulitis treatments with the route of administration as a subfocus, finding that there are only a few low-quality randomized trials comparing the different routes, showing no difference between IV and oral treatments. Cross et al [6], with their primary focus on the duration of therapy, arrived at a similar conclusion. This study hopes to build upon these previous findings, as we have the routes of administration as our primary comparison [6]. We will also include observational studies and place no language barriers in our inclusion criteria, with the hope that it could provide additional valuable data upon Brindle et al’s [3] and Cross et al’s [6] conclusions on oral and IV treatments being equivalent. We hope the results of this study will inform the future prescribing practice of physicians treating cellulitis, especially in deciding oral versus IV treatments. This would not only benefit the patients but also help control the widening spread of antibiotic resistance in the community.