Subscribe for unlimited access to DynaMed content, CME/CE & MOC credit, and email alerts on content you follow.

Already subscribed? Sign in now



General Information


  • cellulitis is acute infection of the dermis and subcutaneous tissue6
  • may be caused by a variety of organisms but most commonly skin flora4,2
    • majority of infections due to streptococcal species in non-purulent cellulitis
    • Staphylococcus aureus less common cause, typically associated with open wound, previous penetrating trauma or purulent infections such as abscesses


  • cellulitis refers to an acute, superficial, diffuse spreading skin infection2
  • definitions of skin and soft tissue infections
    ErysipelasSuperficial infection of the skin involving lymphatics
    CellulitisAcute infection of skin involving deep dermis and subcutaneous fat
    FolliculitisSuperficial infection of hair follicles with purulence in epidermis
    FuruncleHair follicle infection with associated small subcutaneous abscess
    CarbuncleCluster of furuncles
    Cutaneous abscessLocalized collection of pus within dermis and deeper skin tissue
    PyomyositisPurulent infection of skeletal muscle, often with abscess
    ImpetigoSuperficial infection of skin characterized by pustules or vesicles that progress to crusting or bullae
    EcthymaDeeper variant of impetigo evolving into ulcer
    Clostridium myonecrosis (gas gangrene)Necrotizing infection involving muscle
    Necrotizing fasciitisAggressive infection of subcutaneous tissue spreading along fascial planes

    Reference -
                                     29404597Ann Intern Med 2018 Feb 6;168(3):ITC17


  • definition of erysipelas varies2,6
    • some experts distinguish erysipelas from cellulitis based on the depth of infection
      • erysipelas may refer to a more superficial infection involving the upper dermis and superficial lymphatics, with more clearly delineated borders compared to cellulitis
      • cellulitis may refer to infection involving the deeper dermis and adjacent subcutaneous tissue with diffuse, difficult to delineate borders
    • others use erysipelas to refer to cellulitis only involving the face
    • other experts, particularly those in Europe, may use erysipelas and cellulitis interchangeably


General references used

  1. Bystritsky R, Chambers H. Cellulitis and Soft Tissue Infections. Ann Intern Med. 2018 Feb 6;168(3):ITC17-ITC32
  2. Stevens DL, Bisno AL, Chambers HF, et al. Practice guidelines for the diagnosis and management of skin and soft tissue infections: 2014 update by the Infectious Diseases Society of America. Clin Infect Dis. 2014 Jul 15;59(2):e10-52full-text, correction can be found in Clin Infect Dis 2015 May 1;60(9):1448, executive summary can be found in Clin Infect Dis 2014 Jul 15;59(2):147full-text
  3. Mistry RD. Skin and soft tissue infections. Pediatr Clin North Am. 2013 Oct;60(5):1063-82
  4. Liu C, Bayer A, Cosgrove SE, et al. Clinical practice guidelines by the Infectious Diseases Society of America for the treatment of methicillin-resistant Staphylococcus aureus infections in adults and children. Clin Infect Dis. 2011 Feb 1;52(3):e18-e55full-text, correction can be found in Clin Infect Dis 2011 Aug 1;53(3):319, commentary can be found in Clin Infect Dis 2011 Jun 15;52(12):1468, Clin Infect Dis 2011 Jul 1;53(1):97, and in Clin Infect Dis 2011 Aug 1;53(3):308
  5. Stevens DL, Bisno AL, Chambers HF, et al; Infectious Diseases Society of America. Practice guidelines for the diagnosis and management of skin and soft-tissue infections. Clin Infect Dis. 2005 Nov 15;41(10):1373-406full-text, correction can be found in Clin Infect Dis 2005 Dec 15;41(12):1830, Clin Infect Dis 2006 Apr 15;42(8):1219 (dosage error in article text)
  6. Swartz MN. Clinical practice. Cellulitis. N Engl J Med. 2004 Feb 26;350(9):904-12, commentary can be found in N Engl J Med 2004 Jun 10;350(24):2522

Recommendation grading systems used

  • Infectious Diseases Society of America (IDSA) 2014 uses Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system
    • strength of recommendation
      • Strong recommendation - desirable effects clearly outweigh undesirable effects, or vice versa
      • Weak recommendation - desirable effects closely balanced with undesirable effects, or (with low- or very low-quality evidence) uncertainty in estimates of desirable effects, harms, and burden so they may be closely balanced
    • quality of evidence
      • High-quality evidence - consistent evidence from well-performed randomized controlled trials (RCTs) or exceptionally strong evidence from unbiased observational studies
      • Moderate-quality evidence - evidence from RCTs with important limitations (inconsistent results, methodologic flaws, indirect, or imprecise) or exceptionally strong evidence from unbiased observational studies
      • Low-quality evidence - evidence for ≥ 1 critical outcome from observational studies, RCTs with serious flaws, or indirect evidence
      • Very low-quality evidence - evidence for ≥ 1 critical outcome from unsystematic clinical observations or very indirect evidence
  • Infectious Diseases Society of America (IDSA) grading system for recommendations
    • strength of recommendation grades
      • Grade A - good evidence to support recommendation for or against use
      • Grade B - moderate evidence to support recommendation for or against use
      • Grade C - poor evidence to support recommendation
    • quality of evidence ratings
      • I - evidence from ≥ 1 properly randomized, controlled trial
      • II - evidence from ≥ 1 well-designed nonrandomized clinical trial; from cohort or case-controlled analytic studies (preferably from > 1 center); from multiple time series; or from dramatic results from uncontrolled studies
      • III - evidence from opinions of respected authorities, based on clinical experience, descriptive studies, or reports of expert committees
    • Reference - IDSA clinical practice guideline on treatment of methicillin-resistant Staphylococcus aureus infections in adults and children (21208910Clin Infect Dis 2011 Feb;52(3):e18), correction can be found in Clin Infect Dis 2011 Aug 1;53(3):319, commentary can be found in 25609680Clin Infect Dis 2015 Apr 15;60(8):1290

Synthesized Recommendation Grading System for DynaMed Content

  • The DynaMed Team systematically monitors clinical evidence to continuously provide a synthesis of the most valid relevant evidence to support clinical decision-making (see 7-Step Evidence-Based Methodology).
  • Guideline recommendations summarized in the body of a DynaMed topic are provided with the recommendation grading system used in the original guideline(s), and allow users to quickly see where guidelines agree and where guidelines differ from each other and from the current evidence.
  • In DynaMed content, we synthesize the current evidence, current guidelines from leading authorities, and clinical expertise to provide recommendations to support clinical decision-making in the Overview & Recommendations section.
  • We use the Grading of Recommendations Assessment, Development and Evaluation (GRADE) to classify synthesized recommendations as Strong or Weak.
    • Strong recommendations are used when, based on the available evidence, clinicians (without conflicts of interest) consistently have a high degree of confidence that the desirable consequences (health benefits, decreased costs and burdens) outweigh the undesirable consequences (harms, costs, burdens).
    • Weak recommendations are used when, based on the available evidence, clinicians believe that desirable and undesirable consequences are finely balanced, or appreciable uncertainty exists about the magnitude of expected consequences (benefits and harms). Weak recommendations are used when clinicians disagree in judgments of relative benefit and harm, or have limited confidence in their judgments. Weak recommendations are also used when the range of patient values and preferences suggests that informed patients are likely to make different choices.
  • DynaMed synthesized recommendations (in the Overview & Recommendations section) are determined with a systematic methodology:
    • Recommendations are initially drafted by clinical editors (including ≥ 1 with methodological expertise and ≥ 1 with content domain expertise) aware of the best current evidence for benefits and harms, and the recommendations from guidelines.
    • Recommendations are phrased to match the strength of recommendation. Strong recommendations use "should do" phrasing, or phrasing implying an expectation to perform the recommended action for most patients. Weak recommendations use "consider" or "suggested" phrasing.
    • Recommendations are explicitly labeled as Strong recommendations or Weak recommendations when a qualified group has explicitly deliberated on making such a recommendation. Group deliberation may occur during guideline development. When group deliberation occurs through DynaMed Team-initiated groups:
      • Clinical questions will be formulated using the PICO (Population, Intervention, Comparison, Outcome) framework for all outcomes of interest specific to the recommendation to be developed.
      • Systematic searches will be conducted for any clinical questions where systematic searches were not already completed through DynaMed content development.
      • Evidence will be summarized for recommendation panel review including for each outcome, the relative importance of the outcome, the estimated effects comparing intervention and comparison, the sample size, and the overall quality rating for the body of evidence.
      • Recommendation panel members will be selected to include at least 3 members that together have sufficient clinical expertise for the subject(s) pertinent to the recommendation, methodological expertise for the evidence being considered, and experience with guideline development.
      • All recommendation panel members must disclose any potential conflicts of interest (professional, intellectual, and financial), and will not be included for the specific panel if a significant conflict exists for the recommendation in question.
      • Panel members will make Strong recommendations if and only if there is consistent agreement in a high confidence in the likelihood that desirable consequences outweigh undesirable consequences across the majority of expected patient values and preferences. Panel members will make Weak recommendations if there is limited confidence (or inconsistent assessment or dissenting opinions) that desirable consequences outweigh undesirable consequences across the majority of expected patient values and preferences. No recommendation will be made if there is insufficient confidence to make a recommendation.
      • All steps in this process (including evidence summaries which were shared with the panel, and identification of panel members) will be transparent and accessible in support of the recommendation.
    • Recommendations are verified by ≥ 1 editor with methodological expertise, not involved in recommendation drafting or development, with explicit confirmation that Strong recommendations are adequately supported.
    • Recommendations are published only after consensus is established with agreement in phrasing and strength of recommendation by all editors.
    • If consensus cannot be reached then the recommendation can be published with a notation of "dissenting commentary" and the dissenting commentary is included in the topic details.
    • If recommendations are questioned during peer review or post publication by a qualified individual, or reevaluation is warranted based on new information detected through systematic literature surveillance, the recommendation is subject to additional internal review.

DynaMed Editorial Process

  • DynaMed topics are created and maintained by the DynaMed Editorial Team and Process.
  • All editorial team members and reviewers have declared that they have no financial or other competing interests related to this topic, unless otherwise indicated.
  • DynaMed content includes Practice-Changing Updates, with support from our partners, McMaster University and F1000.

Special acknowledgements

  • The American College of Physicians (Marjorie Lazoff, MD, FACP; ACP Deputy Editor, Clinical Decision Resource) provided review in a collaborative effort to ensure DynaMed provides the most valid and clinically relevant information in internal medicine.
  • DynaMed topics are written and edited through the collaborative efforts of the above individuals. Deputy Editors, Section Editors, and Topic Editors are active in clinical or academic medical practice. Recommendations Editors are actively involved in development and/or evaluation of guidelines.
  • Editorial Team role definitions
    Topic Editors define the scope and focus of each topic by formulating a set of clinical questions and suggesting important guidelines, clinical trials, and other data to be addressed within each topic. Topic Editors also serve as consultants for the internal DynaMed Editorial Team during the writing and editing process, and review the final topic drafts prior to publication.
    Section Editors have similar responsibilities to Topic Editors but have a broader role that includes the review of multiple topics, oversight of Topic Editors, and systematic surveillance of the medical literature.
    Recommendations Editors provide explicit review of Overview and Recommendations sections to ensure that all recommendations are sound, supported, and evidence-based. This process is described in "Synthesized Recommendation Grading."
    Deputy Editors oversee DynaMed internal publishing groups. Each is responsible for all content published within that group, including supervising topic development at all stages of the writing and editing process, final review of all topics prior to publication, and direction of an internal team.

How to cite

National Library of Medicine, or "Vancouver style" (International Committee of Medical Journal Editors):

  • DynaMed [Internet]. Ipswich (MA): EBSCO Information Services. 1995 - . Record No. T116794, Cellulitis; [updated 2018 Nov 30, cited place cited date here]. Available from Registration and login required.

Published by EBSCO Information Services. Copyright © 2020, EBSCO Information Services. All rights reserved. No part of this may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission.

EBSCO Information Services accepts no liability for advice or information given herein or errors/omissions in the text. It is merely intended as a general informational overview of the subject for the healthcare professional.


Subscribe for unlimited access to DynaMed content.
Already subscribed? Sign in