The Management of Cellulitis and Erysipelas at an Academic

Emergency Care Journal 2017; volume 13:6343

The management of cellulitis and erysipelas at an academic Introduction Correspondence: Jeff Martin, Department of Visits for cellulitis and erysipelas repre- Family Medicine, Queen’s University, emergency department: current Kingston, 76 Stuart St, Kingston, ON K7L sent a large proportion of cases seen by practice versus the literature 2V7, Canada. emergency physicians. Basic emergency Tel: +613-548-3232. department management consists of antibi- E-mail: [email protected] Jeffrey W. Martin,1 Ruth Wilson,1 otics and appropriate supportive care, deter- Tim Chaplin2 mining if an admission to hospital is neces- Key words: Cellulitis, Antibiotics, Emergency medicine. 1Department of Family Medicine, Queen’s sary, and arranging appropriate follow-up. Antibiotic therapy targeted against beta- University, Kingston, Ontario; Acknowledgements: we thank David Barber hemolytic streptococci and Staphylococcus 2Department of Emergency Medicine, for assisting with electronic medical record aureus, methicillin-sensitive or methicillin- data collection. Queen’s University, Kingston, Ontario, resistant, is the mainstay of treatment for 1 Canada children and adults with these infections. Contributions: JWM contributed to study con- However, overall severity of illness and cept and design, acquisition of the data, analy- underlying comorbidities ultimately deter- sis and interpretation of the data, drafting of mine variables such as delivery route, the manuscript, and critical revision of the dosage, frequency, and class of agent.2 manuscript for important intellectual content. Abstract CRW and TC contributed to study concept and Cephalexin, a first generation design, critical revision of the manuscript for Cellulitis and erysipelas are common cephalosporin, is the preferred oral antibiot- important intellectual content, and study presentations to emergency departments ic for uncomplicated cellulitis without supervision. and family physicians. Evidence-based abscess.3 Published indications for parenter- guidelines for appropriate management of al antibiotics in cellulitis include compro- Conflicts of interest: the authors declare no these infections exist in Canada, but incon- mised oral administration, immunocompro- potential conflict of interest. sistent practices persist. Our objective was mise, and signs of systemic infection.2 Conference presentation: a poster based on These guidelines are incorporated into rec- to determine the level of adherence to cur- this manuscript was accepted for presentation rent evidence and guidelines by emergency ommendations for providers across 4-6 at the Canadian Association of Emergency physicians at the two hospitals in Kingston, Canada. Outpatient parenteral antibiotic Physicians (CAEP) Conference 2017. Ontario, Canada. We identified all of the therapy (OPAT) is chosen for some patients, electronic medical records of patients who and has several advantages, namely as a Received for publication: 11 October 2016. cost-saving measure for the healthcare sys- Revision received: 24 May 2017. were seen at Kingston General Hospital or Accepted for publication: 25 May 2017. Hotel Dieu Hospital between January 1, tem, because patients avoid admission to hospital and repeat visits to the emergency 2015 and June 30, 2015 and given a diagno- This work is licensed under a Creative department.7 In this study, we describe the sis of cellulitis or erysipelas. We randomly Commons Attribution 4.0 License (by-nc 4.0). management practices of emergency physi- selected 182 charts and conducted a retro- cians in Kingston, Ontario for patients diag- ©Copyright J.W. Martin et al., 2017 spective chart review, manually collecting nosed with cellulitis or erysipelas. Licensee PAGEPress, Italy data for patient demographics, medical his- Emergency Care Journal 2017; 13:6343 tory, and medical management. Oral doi:10.4081/ecj.2017.6343 cephalexin alone was given to 44% of our sample, and it was the most common form Materials and Methods of therapy for uncomplicated cellulitis. 36% domly selected a subset of these charts for of patients given any antibiotics at all Study design and population data collection, but excluded charts if they received at least one dose of parenteral We conducted a retrospective chart were inaccessible or did not contain usable antibiotics, despite only 6.7% of these review at Kingston General Hospital data, patients were not seen by an emergen- patients showing systemic signs of illness. (KGH) and Hotel Dieu Hospital (HDH) in cy physician, patients were younger than 1 88% of those receiving parenteral antibi- Kingston, Ontario, Canada. KGH is a ter- year old, or there was a suspected diagnosis tiary care centre and HDH is an ambulatory otics received ceftriaxone, a broad-spec- of preseptal or orbital cellulitis. urgent care centre. They are both teaching trum, third generation cephalosporin. We Patient data were manually collected for hospitals and serve about 500,000 residents found wide variation in antibiotic selection vital signs at time of triage, and patient fac- in southeastern Ontario. A team of emergen- and route of administration for patients pre- tors or comorbidities that may increase the cy physicians staff both the KGH emergen- senting to the emergency department with risk of complications of cellulitis and cy department and the HDH urgent care cellulitis or erysipelas. Overuse of antibi- erysipelas or may predispose to communi- centre and see approximately 100,000 otics is common, and we believe the use of ty-acquired MRSA infection. The comor- patients per year. Ethics approval was parenteral antibiotics may have been unnec- bidities and patient factors we included obtained from the Queen’s University essary for some patients in our sample. Health Sciences Research Ethics Board. were known diabetes, active cancer treat- Emergency physicians should align their ment, documented cancer diagnosis within management plans more closely with the Data collection 6 months, history of organ transplant, intra- current guidelines to improve practice and We identified all of the charts from the venous drug use, or residence in an assisted reduce unnecessary administration of electronic medical record (EMR) for the 6- living environment or shelter. We deter- broad-spectrum parenteral antibiotics. month period between January 1, 2015 and mined whether antibiotics were given; the June 30, 2015 for patients with a discharge dose and route of the antibiotics (parenteral, diagnosis of cellulitis or erysipelas. We ran- topical, oral); whether they were admitted

[page 28] [Emergency Care Journal 2017; 13:6343] Article or discharged home with follow up; dura- tions thereof (n=5). Three patients were prim/sulfamethoxazole, and metronidazole tion and length of time until route switch, noted to be discharged with oral antibiotics, were used alone or in combination with a where applicable; as well as any explicit but no specific details were available. cephalosporin in 22% (12 of 55) of cases, reasoning behind the choice of treatment. Thirteen of those given only oral antibiotics but explicit reasoning was not recorded in We also tracked who returned to the emer- returned to the emergency department for a most of the charts (Figure 2). gency department within two weeks of ini- scheduled recheck or to request reassess- Emergency physicians used a variety of tial presentation for worsening disease, ment. Six of these patients continued their antibiotic classes, combinations, and routes adverse events related to treatment; and existing course of management, but six oth- of administration (Table 2). Four patients whether there were any complications relat- ers received parenteral antibiotics, and one were noted to have allergies to penicillin or ed to treatment, such as allergic reactions. patient started ciprofloxacin and metronida- cephalexin, but clinical reasoning for antibi- zole but was eventually admitted after fur- otic choice and acknowledgment of the Data analysis ther return visits. presence of allergies was otherwise absent Data were analysed using descriptive Greater than one third of the patients in the documentation. statistics. Data abstraction and analysis given any antibiotics at their first visit (60 There were 62 return visits to the were done using Microsoft Excel. of 166) received empiric parenteral therapy emergency department within 2 weeks of for cellulitis, then were switched to or aug- discharge from emergency. This number mented with oral antibiotics if improvement included 46 who returned for scheduled was observed. 88% (53 of 60) received cef- rechecks, repeat doses of IV antibiotics; or Results triaxone, some augmented with van- addition of IV antibiotics to the manage- comycin (n=3) or metronidazole (n=2), and ment plan, due to worsening infection Patients’ characteristics the remaining patients received cefazolin There were 707 visits to KGH and HDH (n=5 [one patient received both ceftriaxone (n=3). Nine patients were admitted upon for cellulitis or erysipelas between January and cefazolin]), clindamycin (n=1), cef- returning for rechecks. Of the remaining 2015 and June 2015. We randomly selected tazidime (n=1), or an unknown parenteral visits, one patient returned with anaphy- 182 of these charts for study, 178 of which antibiotic (n=1) (Figure 1). One patient laxis, possibly secondary to the cephalexin were for patients diagnosed with cellulitis given parenteral antibiotics during their ini- or doxycycline prescribed; a second and four of which were for patients with tial visit was also admitted. Four of the 60 patient returned with a possible drug rash erysipelas. Patient characteristics for our patients (6.7%) exhibited tachycardia or after receiving two intravenous doses of sample population are presented in Table 1. fever at triage. Six patients were sent home ceftriaxone and oral trimethoprim/sul- Four charts selected in the process were for with supports set up to administer OPAT, famethoxazole; and a third patient needed return visits to the emergency department for although one was subsequently admitted. an indwelling IV line removed and rein- worsening disease or complications, or for Of the 55 patients with one or more serted. The other return visits by patients scheduled reassessment. We collected data comorbidities as listed above, all received to the emergency department (n=13) were from these records but also from the patient’s antibiotics. 44% (24 of 55) received for separate reasons unrelated to the initial initial presentation for the infection. cephalexin only. Clindamycin, trimetho- diagnosis. Management plans In 16 of 182 visits, patients received no Table 1. Patients’ characteristics. antibiotics (Table 1; Figure 1). One of these patients was a 63-year-old man with normal Characteristics N = 182 vital signs who had an incision and drainage Median age (range), years 49 (2-94) performed for an abscess at his initial visit, Sex, male/female, n (%) 103 (57) / 79 (43) but returned the next day with cellulitis. This man had normal vital signs at his Institution visited KGH 66 (36) return visit and received two doses of intra- HDH 116 (64) venous ceftriaxone 24 hours apart before starting a course of cephalexin. The remain- Vital signs at triage Fever*, n (%) 6 (3) ing fifteen patients did not return to the Tachycardia°, n (%) 30 (16) emergency department within the following two weeks. 106 patients received oral Comorbidities Diabetes, type I or II, n (%) 31 (17) antibiotics only or were advised to continue End-stage renal disease, n (%) 3 (2) existing courses of oral antibiotics (Table 1; Organ transplant#, n (%) 1 (0.5) Figure 1). Although cephalexin was the sole Concurrent malignancy, n (%) 6 (3) agent prescribed in 72 of those cases, seven HIV/AIDS, n (%) 1 (0.5) patients received combinations with Intravenous drug use, n (%) 13 (7) trimethoprim/sulfamethoxazole (n=1), Living in assisted environment or shelter, n (%) 10 (5) ciprofloxacin (n=2), doxycycline (n=1), or History of MRSA infection, n (%) 5 (3) metronidazole (n=3). In those not receiving Initial antibiotic therapy administered any cephalexin initially, initial agents were No antibiotics, n (%) 16 (9) clindamycin (n=7), trimethoprim/sul- Oral antibiotics only, n (%) 106 (58) famethoxazole (n=4), ciprofloxacin (n=3), Empiric parenteral antibiotics, n (%) 60 (33) amoxicillin-clavulanate (n=1; for a cat bite *Oral temperature >38.0°C; °heart rate greater than 160 beats per minute (bpm) for infants (<2 years), greater than 140 bpm; for children victim), or doxycycline (n=4; for suspected (2 to 12 years), and greater than 100 bpm for adolescents and adults; #the only patient with an organ transplant had had a renal transplant. early localized Lyme disease), or combina-

[Emergency Care Journal 2017; 13:6343] [page 29] Article

systemic involvement to minimize treat- rors a previous review of emergency depart- Discussion ment failure.10 However, there should not be ment management of cellulitis in the 13 Cellulitis and erysipelas are common overreliance on parenteral therapy when province of Alberta, Canada. However, reasons for emergency department and there is no evidence of improved efficacy. they describe a significantly higher use of urgent care visits in our population. In fact, Overuse of antibiotics in general, intravenous cefazolin (47% of initial treat- within our study period of six months, we though not a focus of this study, has other ment regimens) versus oral cephalexin (8% identified over 700 visits related to the skin risks. Two patients returned after suspected of initial treatment regimens). In our data infections. Outpatient management with allergic response to antibiotics, including set, oral cephalexin was given in 44% of the self-administration of oral cephalexin was one with anaphylaxis. Allergic reaction to initial treatment plans versus intravenous the most common choice of management antibiotics is not uncommon11 and it rein- cefazolin in only 2.7% of the initial treat- for cellulitis and erysipelas. This is the rec- forces the importance of justification for ment plans. Ceftriaxone, a broad-spectrum ommended antibiotic for uncomplicated antibiotics whenever they are prescribed. third generation cephalosporin with good cellulitis considering its spectrum of activi- There is no widely used infection sever- coverage of Gram-negative bacteria but less 2,3 ty. We note more than one third of the ity scale in Canada but evidence-based activity against Gram-positive bacteria than patients in our study given any antibiotics sources do exist to help inform treatment earlier generation cephalosporins, was the were initially treated with empiric parenter- plans.12 We found substantial variation most frequently used parenteral antibiotic in al therapy for cellulitis then switched to or among emergency physicians in manage- our study. Current guidelines reserve ceftri- continued on oral antibiotics if improvement of cellulitis and erysipelas. This mir- axone for severe, deep infections involving ment was noted after 48 to 72 hours. However, almost all of these patients were systemically well at the time of their triage assessment. Canadian expert guidelines dic- Table 2. Type/route of antibiotics received for cellulitis/erysipelas at the initial visit to the emergency department. tate conservative management with oral medication, with the addition of parenteral Type Number (% of 182 total) medication if there are signs of worsening Cephalexin PO 80 (44.0) disease despite two or more days of oral therapy.4,5 Ceftriaxone IV, cephalexin PO 30 (16.5) Skin infections are largely clinical diag- Clindamycin PO 6 (3.3) noses, and in the absence of systemic signs Ceftriaxone IV/IM 4 (2.2) of illness, clinical improvement in cellulitis Ceftriaxone IV, Septra PO 4 (2.2) and erysipelas should be monitored during Ciprofloxacin PO 4 (2.2) the 48 to 72 hour period after initiation of Septra PO 3 (1.6) treatment. Of those patients in our study population who returned to the emergency Cefazolin IV 3 (1.6) department reassessment after starting an Ceftriaxone IV, clavulin PO 3 (1.6) oral agent, just less than half received par- Cephalexin PO and septra PO 2 (1.1) enteral antibiotics despite normal vital Cephalexin PO, doxycycline PO 2 (1.1) signs. It is difficult for us to comment on Amoxicillin PO 2 (1.1) whether parenteral antibiotics were given for worsening clinical appearance because Clarithromycin PO 2 (1.1) of limited detail in charting. However, in Ceftriaxone IV, vancomycin IV, septra PO 2 (1.1) the absence of abnormal vital signs and Cefazolin IV, Cephalexin PO 2 (1.1) given the erythema and edema of cellulitis Clindamycin IV, Clindamycin PO 1 (0.5) may worsen in the first 24 to 48 hours of Ciprofloxacin PO, Clindamycin PO 1 (0.5) treatment,4,7 we question whether a switch Amoxicillin-clavulinate PO 1 (0.5) to or augmentation with parenteral antibiotics was necessary in all cases. Doxycycline PO and metronidazole PO 1 (0.5) Overuse of parenteral antibiotics is very Cephalexin PO and metronidazole PO 1 (0.5) common,8 but a recent Cochrane review Ceftriaxone IV, Cloxacillin PO 1 (0.5) suggests oral antibiotics (macrolides/strep- Ceftriaxone IV, Cefazolin IV, Cephalexin PO 1 (0.5) togramins) may be more effective for cel- Ceftriaxone IV, Metronidazole IV, Cephalexin PO 1 (0.5) lulitis and erysipelas than penicillin given parenterally and recommends further study Ceftriaxone IV, cephalexin PO, valacyclovir PO 1 (0.5) on this subject.9 We must assume physicians Ceftriaxone IV, vancomycin IV 1 (0.5) seeing patients included in our study chose Ceftriaxone IV, Penicillin V PO 1 (0.5) parenteral administration based on gross Doxycycline PO 1 (0.5) appearance of the cellulitis possibly in com- Ceftazidime IV, Cefprozil PO 1 (0.5) bination with other patient factors, such as Vancomycin IV 1 (0.5) poor reliability. Such extenuating patient factors, if present, were not described in the Ceftriaxone IM, Cephalexin PO, Ciprofloxacin PO 1 (0.5) chart records, so we cannot accurately anal- Ceftriaxone IV, Cephalexin PO, fluconazole PO 1 (0.5) yse decision-making. Management of these Unknown antibiotic IV, Cephalexin PO 1 (0.5) infections should vary depending on the No antibiotics 16 (8.8) presence of chronic illnesses and signs of

[page 30] [Emergency Care Journal 2017; 13:6343] Article other tissues and organs; and promote cefa- Nova Scotia Adult Cellulitis Guidelines, zolin intravenously for mild, moderate, or developed in 2000 and validated in subse- Conclusions 2,4-6 6 severe cellulitis. However, some advo- quent studies; Toronto’s University Health Our study has found a wide variety in cate for ceftriaxone given its smaller side Network;4 and Providence Health Care in the antibiotic selection and route of admin- effect profile and ease of administration.7 In British Columbia.5 The guidelines are large- istration for the treatment of cellulitis and the United Kingdom, the narrow-spectrum ly consistent with each other, as well as with penicillin flucloxacillin, which is available other international guidelines,2,12,14 and erysipelas in patients presenting to two ter- orally or intravenously, is suggested ahead accurately reflect published evidence in the tiary care emergency departments in the of broader-spectrum agents.14 Regarding infectious disease literature. In the present province of Ontario. Outpatient administra- OPAT and return visits to the emergency study, we were unable to determine if emer- tion of oral cephalexin is the predominant department for repeat parenteral dosing, gency physicians adhered to any of these form of therapy for cellulitis and erysipelas ceftriaxone may be more sensible given its rules, and documentation in the EMR would and this is consistent with management once daily administration. However, once- suggest they were not. Limitations to this guidelines. However, a significant number daily cefazolin in combination with oral study include the small sample size. We of patients received parenteral broad-spec- probenecid is equivalent to ceftriaxone in used random sampling to collect 182 charts trum antibiotics and continued on this as 15 for this chart review, because our aim was treating moderate/severe cellulitis. outpatients, returning to the emergency A significant limitation to managing to capture a snapshot of antibiotic use in our department for repeat dosing. This would cellulitis effectively is the lack of standard- sample population. Future studies could contribute considerable cost in terms of ized scales of disease severity. No consis- make use of systematic sampling to allow tent explicit reasoning for choice of man- for a larger, more representative sample money, time, and resources to an already agement plan was demonstrated in the over the course of years rather than several busy emergency medicine system. It may be charting by emergency physicians analysed months. Additionally, we were limited by easier and more economically feasible to in this study, and it was not apparent the amount of information documented. administer a once daily broad-spectrum whether physicians were using any grading Charting by emergency physicians was antibiotic, like ceftriaxone. However, we system or algorithm. The majority of brief, as is customary; and pertinent fea- question whether it should be at the expense patients were systemically well according tures, such as appearance of the region of of existing evidence when we know the cellulitis or details about risk factors com- to vital signs taken by the triage nurse, and most likely culprit organism, and when mon in this part of the province such as physicians’ notes demonstrated a lack of newer antimicrobials have near-equivalent subjective constitutional symptoms in the intravenous drug use, were often not record- intravenous and oral bioavailability. The majority of cases when details were docu- ed. We do note that physicians sometimes prevalence of antibiotic resistance is mented. Some Canadian centres have devel- used templates for information gathering, oped their own guidelines, including the but even in those cases, recording of risk increasing and evidence-based prescribing factors such as intravenous drug use was and more appropriate antibiotic usage frequently incomplete. should become increasingly important.

Figure 1. Proportions of patients by route of antibiotic administration. The majority of patients receiving parenteral antibiotic were given ceftriaxone. The other antibi- Figure 2. Numbers of patients by antibiotic regimen received and comorbidities. A wide otics given were given vancomycin, variety of antibiotics were chosen for initial management of patients with cellulitis. Little metronidazole, or ceftazidime. Of those consistency was seen with respect to comorbidities. FQ, fluoroquinolones; PEN, peni- given any intravenous antibiotic, 6.7% had cillins; TET, tetracyclines; MAC, macrolides; MTZ, metronidazole; CLI, clindamycin; tachycardia and fever. SXT, trimethoprim/sulfamethoxazole.

[Emergency Care Journal 2017; 13:6343] [page 31] Article

soft tissue infections. Providence emergency department patients with References HealthCare Emergency Department; uncomplicated cellulitis. Acad Emerg 1. Lopez FA, Lartchenko S. Skin and soft 2010. Med 2014;21:526-31. tissue infections. Infect Dis Clin North 6. Campbell SG, Burton-MacLeod R, 11. Shehab N, Patel PR, Srinivasan A, Am 2006;20:759-72. Howlett T. A cellulitis guideline at a Budnitz DS. Emergency department 2. Stevens DL, Bisno AL, Chambers HF, community hospital – we can reduce visits for antibiotic-associated adverse et al. Practice guidelines for the diagno- costs by standardizing care. J Emerg events. Clin Infect Dis 2008;47:735-43. sis and management of skin and soft tis- Primary Health Care 2009;7:1. 12. Anti-infective guidelines for communi- sue infections: 2014 update by the 7. Eron LJ, Lipsky BA, Low DE, et al. ty-acquired infections. Toronto: MUMS Infectious Diseases Society of America. Managing skin and soft tissue infec- Guideline Clearinghouse; 2013. Clin Infect Dis 2014;59:e10-52. tions: expert panel recommendations on 13. Dong SL, Kelly KD, Oland RC, et al. 3. Pallin DJ, Binder WD, Allen MB, et al. key decision points. J Antimicrob ED management of cellulitis: a review Clinical trial: comparative effectiveness Chemother 2003;52(Suppl 1):3-17. of five urban centers. Am J Emerg Med of cephalexin plus trimethoprim-sul- 8. Waldrop RD, Prejean C, Singleton R. 2001;19:535-40. famethoxazole versus cephalexin alone Overuse of parenteral antibiotics for 14. Guidelines on the Management of for treatment of uncomplicated celluli- wound care in an urban emergency Cellulitis in Adults. Clinical Resource tis: a randomized controlled trial. Clin department. Am J Emerg Med Efficiency Support Team (CREST); Infect Dis 2013;56:1754-62. 1998;16:343-5. 2005. 4. Management of Uncomplicated Skin 9. Kilburn SA, Featherstone P, Higgins B, 15. Grayson ML, McDonald M, Gibson K, and Skin Structure Infections - Non- Brindle R. Interventions for cellulitis et al. Once-daily intravenous cefazolin Purulent Cellulitis and Erysipelas. and erysipelas. Cochrane Database Syst plus oral probenecid is equivalent to Mount Sinai Hospital-University Rev 2010;6:CD004299. once-daily intravenous ceftriaxone plus Health Network Antimicrobial 10. Peterson D, McLeod S, Woolfrey K, oral placebo for the treatment of moder- Stewardship Program; 2014. McRae A. Predictors of failure of ate-to-severe cellulitis in adults. Clin 5. Management of mild/moderate skin and empiric outpatient antibiotic therapy in Infect Dis 2002;34:1440-8.

[page 32] [Emergency Care Journal 2017; 13:6343]