Clinical and laboratory factors associated with prolonged hospital stay among patients with cellulitis/erysipelas

ARTIGO ORIGINAL Re is a Cien í ica da O dem dos Médicos www.ac amedicapo uguesa.com 448 ABSTRACT In oduc ion: Celluli is and e ysipelas ep esen he mos equen cause o hospi aliza ion in he de ma ology depa men o San a Ma ia Hospi al in Lisbon, Po ugal. The aim o his s udy was o in es iga e whe he pa ien demog aphics, como bidi ies, p e ious episodes o celluli is/e ysipelas, he p esence o complica ions, labo a o y ma ke s a admission, mic obial isola ion o p e ious use o an ibio ics, a e associa ed wi h p olonged s ays. Ma e ial and Me hods: Re ospec i e analysis, including pa ien s admi ed wi h celluli is/e ysipelas in he inpa ien de ma ology de- pa men o San a Ma ia Hospi al be ween July 1s 2012 and June 30 h 2017. Resul s: The e we e 372 admissions, co esponding o 348 pa ien s. The median leng h o s ay was 11 days. Inc eased age (p = 0.002, OR 1.03, 95% CI 1.01 – 1.04), p e ious episode o celluli is/e ysipelas equi ing hospi aliza ion (p = 0.005, OR 4.81, 95% CI 1.63 – 14.23), he p esence o celluli is/e ysipelas-associa ed complica ions (p = 0.001, OR 3.28, 95% CI 1.63 – 6.59), leukocy- osis (p = 0.049, OR 1.81, 95% CI 1.00 – 3.30), high le els o C- eac i e p o ein (p = 0.035, OR 1.03, 95% CI 1.00 – 1.06) and a posi i e cul u e esul (p = 0.002, OR 2.59, 95% CI 1.41 – 4.79) we e associa ed wi h p olonged hospi aliza ion. Discussion: P olonged hospi aliza ion o celluli is/e ysipelas is associa ed wi h highe cos s, addi ional clinical in es iga ion, in asi e ea men s, p olonged cou ses o an ibio ic he apy, isk o nosocomial in ec ions, and delayed e u n o ac i i ies o daily li ing. Thus, he in es iga ion o clinical-labo a o y ac o s associa ed wi h p olonged hospi aliza ion o celluli is / e ysipelas is essen ial and may be use ul o he cons uc ion o a se e i y sco e. Conclusion: The knowledge o he cha ac e is ics ha a e associa ed wi h p olonged s ay among pa ien s wi h celluli is/e ysipelas may be ele an o imp o e heal h ca e, by educing he leng h o hospi al s ay and associa ed isks and cos s. Keywo ds: C-Reac i e P o ein; Celluli is; E ysipelas; Hospi aliza ion; Leukocy osis Clinical and Labo a o y Fac o s Associa ed wi h P olonged Hospi al S ay among Pa ien s wi h Celluli is/E ysipelas Fa o es Clínico-Labo a o iais Associados ao In e namen o P olongado em Doen es com Celuli e/E isipela Ângela RODA1, Ana Ma cos PINTO1, Ana Ri a FILIPE2, Ana Ri a TRAVASSOS1, João Ped o FREITAS1, Paulo FILIPE1,3,4 Ac a Med Po 2019 Jun;32(6):448–452 ▪ h ps://doi.o g/10.20344/amp.10735 1. Se iço de De ma ologia. Hospi al de San a Ma ia. Cen o Hospi ala Uni e si á io Lisboa No e. Lisboa. Po ugal. 2. Unidade de Saúde Pública de Oei as. Ag upamen o de Cen os de Saúde de Lisboa Ociden al e Oei as. Oei as. Po ugal. 3. Clínica Uni e si á ia de De ma ologia. Faculdade de Medicina. Uni e sidade de Lisboa. Lisboa. Po ugal. 4. Ins i u o de Medicina Molecula . Faculdade de Medicina. Uni e sidade de Lisboa. Lisboa. Po ugal.  Au o co esponden e: Ângela Roda. [email p o ec ed] Recebido: 28 de ab il de 2018 – Acei e: 02 de ou ub o de 2018 | Copy igh © O dem dos Médicos 2019 RESUMO In odução: A celuli e e a e isipela cons i uem a causa mais equen e de in e namen o no Se iço de De ma ologia do Hospi al San a Ma ia. Es e es udo e e como obje i o in es iga se as ca ac e ís icas demog á icas, as como bilidades, a exis ência de episódios p é ios de celuli e/e isipela, a p esença de complicações associadas, os pa âme os labo a o iais na admissão, o isolamen o de mic o ganismo em cul u a ou o uso p é io de an ibió icos es ão associados a in e namen os p olongados. Ma e ial e Mé odos: Es udo e ospe i o, incluindo os doen es in e nados no Se iço de De ma ologia do Hospi al San a Ma ia com o diagnós ico de celuli e/e isipela, en e 1 de julho de 2012 e 30 de junho de 2017. Resul ados: Exis i am 372 in e namen os, co espondendo a 348 doen es. A mediana do empo de in e namen o oi de 11 dias. A idade (p = 0,002, OR 1,03, 95% IC 1,01 – 1,04), a exis ência de in e namen o p é io po celuli e/e isipela (p = 0,005, OR 4,81, 95% IC 1,63 – 14,23), a p esença de complicações associadas à celuli e/e isipela (p = 0,001, OR 3,28, 95% IC 1,63 – 6,59), a leuco- ci ose (p = 0,049, OR 1,81, 95% IC 1,00 – 3,30), alo es ele ados de p o eína C ea i a (p = 0,035, OR 1,03, 95% IC 1,00 - 1,06) e o isolamen o de mic o ganismo em cul u a (p = 0,002, OR 2,59, 95% IC 1,41 – 4,79) es i e am associados a in e namen os p olongados. Discussão: A pa dos maio es cus os associados, o in e namen o p olongado po celuli e/e isipela es á equen emen e associado à necessidade de in es igação clínica adicional, a a amen os in asi os, a cu sos p olongados de an ibio e apia, ao isco de in eções nosocomiais e ao a aso no e o no às a i idades da ida diá ia. Assim, o es udo dos a o es clínico-labo a o iais associados ao in e - namen o p olongado po celuli e/e isipela é undamen al e pode á se ú il pa a a cons ução de um sco e de g a idade. Conclusão: O conhecimen o de ca ac e ís icas clínicas e labo a o iais associadas ao in e namen o p olongado pode á se ele an e pa a melho a os cuidados de saúde, a a és da edução dos empos de in e namen o e dos seus iscos e cus os associados. Pala as-cha e: Celuli e; E isipela; Hospi alização; Leucoci ose; P o eína C-Reac i a INTRODUCTION A Celluli is and e ysipelas a e he leading causes o admission o he Depa men o De ma ology o he Hos- pi al San a Ma ia, co esponding o a ound 18% o annual admissions. E en hough celluli is can be di e en ia ed om e ysip- elas, his is mainly based on unclea clinical c i e ia and is i ele an ega ding he app oach o ea men and man- agemen . The e o e, bo h e ms a e in e changeably used in clinical p ac ice.1 Up o da e, li le is known on he ac o s wi h an in luence ARTIGO ORIGINAL Re is a Cien í ica da O dem dos Médicos www.ac amedicapo uguesa.com 449 Roda A, e al. Fac o s associa ed wi h p olonged hospi al s ay among pa ien s wi h celluli is/e ysipelas, Ac a Med Po 2019 Jun;32(6):448–452 on hospi al leng h o s ay (LOS) o pa ien s wi h celluli is/e - ysipelas. This s udy was aimed a de ining he associa ion be ween p olonged LOS and pa ien ’s demog aphic cha ac- e is ics, como bidi ies, he p esence o p e ious episodes o celluli is/e ysipelas o p e ious complica ions associa ed wi h bo h condi ions, labo a o y pa ame e s on admission, posi i e cul u es o p e ious an ibio ic ea men . MATERIAL AND METHOD This was a e ospec i e s udy in ol ing pa ien s aged 18 o olde admi ed o he Depa men o De ma ology o he Hospi al de San a Ma ia be ween 1 Jul 2012 and 30 Jun 2017 and diagnosed wi h celluli is / e ysipelas, ega dless o he loca ion o he a ec ed a ea. The ollowing explana o y a iables we e assessed: de- mog aphic cha ac e is ics, including pa ien ’s age and gen- de , he p esence o como bidi ies, namely diabe es melli- us, pe iphe al ascula disease (a ec ing eins, a e ies and lympha ic essels), immune impai men (associa ed wi h in ec ious, haema ological and pha macological dis- ease); his o y o celluli is/e ysipelas, namely he p esence o a leas one p e ious episode o celluli is/e ysipelas, a leas one p e ious admission due o celluli is/e ysipelas and compliance wi h an ibio ic p ophylaxis; he p esence o complica ions associa ed wi h celluli is/e ysipelas (in- cluding abscess, lymphangi is, nec osis, skin ulce s, blis- e s, os eoa icula in ol emen ); he p esence o a po al o en y [all he condi ions in ol ing skin ba ie dis up ion, including ungal in ec ions o he skin, hai and nails, i al (he pe ic in ec ions), bac e ial ( olliculi is, boils), auma ic wounds, su gical wounds, insec bi es and s ings]; labo- a o y pa ame e s on admission, namely whi e blood cell coun , neu ophil coun , C- eac i e p o ein (CRP) le el; posi i e cul u e (ulce o abscess exuda e and/o blood cul- u e); an ibio ic ea men o he cu en episode o celluli is/ e ysipelas on he week p e ious o admission. The selec ion o a iables was based on he pa hophys- iology o he disease and i s hypo he ical ela ionship wi h he leng h o s ay in hospi al. S a is ical analysis Ca ego ical a iables we e p esen ed in he desc ip i e analysis as equencies and a es and con inuous a iables as means and s anda d de ia ions, o medians and in e - qua ile ange, depending on he ype o da a (no mally o non-no mal dis ibu ed, espec i ely). The Shapi o-Wilk es o no mali y was used and cu -o based [-1; 1] skewness and ku osis alues we e analysed. Hospi al admissions we e di ided in o wo di e en g oups, depending on he leng h o s ay (g ea e han 14 days o 14 days and less); his dicho omisa ion was based on he median LOS o ou g oup o pa ien s and was sub- sequen ly de ined. A leng h o hospi al s ay g ea e han 14 days was conside ed as a p olonged LOS. Chi-squa e o Fishe ’s exac es we e used in bi a i- a e analysis in o de o check o an associa ion be ween ca ego ical a iables and S uden ’s - es was used o he assessmen o con inuous a iables. Finally, an imp o ed logis ic eg ession model was de- eloped, wi h a s epwise da a en y me hod, including s a- is ically signi ican a iables in bi a ia e analysis, aimed a de ining which a iables we e associa ed wi h p olonged LOS. Only wo- ailed p- alues we e epo ed, wi h a 0.05 sig- ni icance le el (α). Da a analysis was made by use o he SPSS, e sion 23 so wa e. Table 1 – Uni a ia e analysis o he cha ac e is ics o pa ien s admi ed wi h celluli is/e ysipelas Gene al (n = 372) Sho s ay (n = 247) P olonged s ay (n = 125) p- alue Age – mean (SD) (yea s) 61.2 (18.4) 59.5 (18.7) 64.5 (17.4) 0.01 Female gende – n (%) 191 (51%) 122 (49%) 69 (55%) 0.29 Diabe es melli us – n (%) 86 (23%) 50 (20%) 36 (29%) 0.06 Pe iphe al ascula disease – n (%) 96 (26%) 55 (22%) 41 (33%) 0.03 Immune impai men – n (%) 47 (13%) 29 (12%) 18 (14%) 0.47 P e ious local su ge y – n (%) 83 (22%) 51 (21%) 32 (26%) 0.28 P e ious episode(s) o celluli is/e ysipelas – n (%) 79 (21%) 50 (20%) 29 (23%) 0.51 P e ious hospi al admission due o celluli is/e ysipelas – n (%) 24 (6%) 8 (3%) 16 (13%) < 0.001 An ibio ic p ophylaxis 15 (4%) 9 (4%) 6 (5%) 0.59 Complica ions – n (%) 66 (18%) 32 (13%) 34 (27%) 0.001 De ined po al o en y – n (%) 193 (52%) 129 (52%) 64 (51%) 0.85 Leukocy osis – n (%) 188 (51%) 109 (44%) 79 (63%) 0.001 Neu ophilia – n (%) 221 (59%) 135 (55%) 86 (69%) 0.009 CRP – mean (SD) (mg/dL) 12.4 (9.9) 11.2 (8.9) 14.8 (11.5) 0.002 Posi i e cul u e* – n (%) 82 (22%) 37 (15%) 45 (36%) < 0.001 An ibio ic ea men on he week p e ious o admission – n (%) 111 (30%) 67 (27%) 44 (35%) 0.11 ARTIGO ORIGINAL Re is a Cien í ica da O dem dos Médicos www.ac amedicapo uguesa.com 450 Roda A, e al. Fac o s associa ed wi h p olonged hospi al s ay among pa ien s wi h celluli is/e ysipelas, Ac a Med Po 2019 Jun;32(6):448–452 RESULTS A o al o 372 hospi al admissions we e eco ded du ing he s udy pe iod, co esponding o 348 pa ien s, wi h an 11- day median leng h o s ay, mos ly a ec ing he lowe limb (n = 284, 76%), ollowed by he ace (n = 42, 11%) and he uppe limb (n = 36, 10%). Fo y ou pe cen o he hospi al admissions we e conside ed as p olonged LOS (n = 125). Pa ien ’s cha ac e is ics a e shown in Table 1. A signi i- can ly highe mean age was ound in he g oup o pa ien s wi h p olonged LOS [64.5 (17.4) s. 59.5 (18.7); p = 0.01] while no signi ican di e ences we e ound as ega ds pa- ien ’s gende (p = 0.29), e en hough mos ly emale pa- ien s wi h p olonged LOS we e ound. Ra es o 29% s. 20%, 33% s. 22%, 14% s. 12% and 26% s. 21% we e ound as ega ds p olonged s. sho s ay, espec i ely, in pa ien s wi h diabe es melli us, pe iphe al ascula disease, immune impai men and p e ious local su ge y. A p e ious his o y o celluli is/e ysipelas was ound in 21% o hospi al admissions and a leas one p e ious hospi al admission due o he same eason was ound in 6% o he cases. As ega ds he episode o celluli is/e ysipelas unde lying he hospi al admission, he p esence o complica ions was signi ican ly mo e equen in pa ien s wi h p olonged LOS (27% s. 13%; p = 0.001). The complica ions associa ed wi h celluli is/e ysipelas a e shown in Table 2. The iden i ica- ion o a po al o en y was simila in bo h g oups (52% s. 51%). Cul u es we e ca ied ou in 262 admissions, which we e posi i e in 31% (n = 82) o he cases (72 cases wi h a posi i e exuda e cul u e, se en wi h a posi i e blood cul u e and h ee wi h bo h posi i e exuda e and blood cul u es). The mic oo ganisms ha we e g own a e shown in Fig. 1. Finally, no signi ican di e ences we e ound as ega ds an an ibio ic ea men on he week p e ious o admission, e en hough his was mo e equen ly ound in he g oup wi h a p olonged LOS (35% s. 27%) (p = 0.11). The adjus men o he model o logis ic eg ession has allowed o he iden i ica ion o pa ien ’s age (p = 0.002), p e ious admission wi h celluli is/e ysipelas (p = 0.005), p esence o complica ions (p = 0.001), leukocy osis (p = 0.049), CRP alue (p = 0.035) and posi i e cul u e (p = 0.002) as ac o s associa ed wi h p olonged LOS (Ta- ble 3). The p esence o p e ious admission wi h celluli is/ Table 2 – Complica ions associa ed wi h celluli is/e ysipelas Gene al (n = 66) Sho s ay (n = 32) P olonged s ay (n = 34) Abscess – n (%) 33 (50%) 11 (35%) 22 (64%) Lymphangi is – n (%) 22 (33%) 18 (56%) 4 (12%) Skin ulce – n (%) 4 (6%) 1 (3%) 3 (9%) Nec osis – n (%) 1 (2%) 0 (0%) 1 (3%) Skin blis e – n (%) 2 (3%) 1 (3%) 1 (3%) Os eoa icula in ol emen – n (%) 4 (6%) 1 (3%) 3 (9%) Figu e 1 – Mic oo ganisms g own in cul u e. MSSA: me hicillin-sensi i e S aphylococcus au eus; MRSA: me hicillin- esis an S aphylococcus au eus; P. ae uginosa: Pseudomona ae uginosa; spp.: species. O he s, including: Se a ia ma cescens (n = 3), E. coli (n = 1), Klebsiella pneumoniae (n = 1), Pan oea agglo- me ans (n = 1) P olonged s ay Sho s ay MSSA + P. aue ginosa O he P. ae uginosa S ep ococcus spp. MRSA MSSA 2 4 3 1 8 3 7 9 4 16 20 5 ARTIGO ORIGINAL Re is a Cien í ica da O dem dos Médicos www.ac amedicapo uguesa.com 451 e ysipelas and he p esence o complica ions we e he mos signi ican a iables. The au ho s decla e ha ing ollowed he ongoing p o- ocols a he heal hca e ins i u ion as ega ds he publica- ion o any pe sonal da a. An anonymised da ase has been used and none o he au ho s had any access o any iden i- ica ion elemen s o he pa ien s. The s udy was ca ied ou acco ding o he Helsinki Decla a ion o he Wo ld Medical Associa ion. DISCUSSION This is he i s Po uguese s udy on he clinical and lab- o a o ial ac o s associa ed wi h p olonged LOS o pa ien s admi ed wi h celluli is/e ysipelas. The e o e, i was ound ha pa ien ’s age, he p esence o a leas one p e ious hospi al admission wi h celluli is/e ysipelas, he p esence o complica ions, leukocy osis, CRP alue on admission and a posi i e cul u e we e associa ed wi h a long hospi al s ay. In addi ion o highe associa ed cos s, p olonged LOS due o celluli is/e ysipelas is equen ly associa ed wi h he need o addi ional clinical e alua ion wi h diagnos ic es s (including imaging), wi h in asi e p ocedu es, including su gical d ainage o deb idemen , long- e m an ibio ic ea - men , wi h he isk o nosocomial in ec ions and delayed e- co e y o he ac i i ies o daily li ing, wi h an impac on he pa ien ’s quali y o li e. Despi e no ideal an ibio ic egimen has e e been es- ablished, a he apeu ic p o ocol based on he p esence o como bidi ies and p e ious an ibio ic ea men is ollowed in ou depa men . E en hough clinical guidelines ha e al- eady been ecommended o he app oach and ea men o pa ien s wi h celluli is/e ysipelas, he e a e s ill ew da a allowing o he iden i ica ion o pa ien s a isk o p olonged LOS and, in addi ion, ha could bene i om a mo e agg es- si e ea men .2,3 In line wi h ou s udy, o he s udies ha e looked o he cha ac e is ics o pa ien s wi h celluli is/e ysipelas associ- a ed wi h p olonged LOS. Acco ding o he No h-Ame ican mul i-cen ic s udy by Gang e al. on 4,224 hospi al admis- sions due o lowe limb celluli is/e ysipelas, a iables such as pa ien ’s age, emale gende , diabe es melli us, achy- ca dia, hypo ension, leukocy osis, neu ophilia and ele a - ed se um c ea inine we e associa ed wi h p olonged LOS.4 In a s udy om New Zealand, in ol ing 51 pa ien s admi - ed wi h lowe limb celluli is, i has been ound ha an ele- a ed neu ophil coun , he sco e o oedema and he use o diu e ics we e independen ly associa ed wi h p olonged LOS.5 In 2003, Ca a alà e al. ha e examined 332 pa ien s admi ed wi h celluli is, ega dless o i s loca ion and ha e eached he conclusion ha pa ien s wi h mul iple como - bidi ies, hypoalbuminaemia, kidney ailu e o skin nec osis on admission emained longe in he hospi al.6 In addi ion, an Aus alian s udy in ol ed 395 episodes o celluli is/e - ysipelas a ec ing di e en ana omical loca ions has ound ha o e 60 yea s o age, symp oms las ing mo e han 4 days, he p esence o hypoalbuminaemia, he p esence o bac e aemia and me hicillin- esis an S aphylococcus au eus (MRSA) we e associa ed wi h p olonged LOS. On he o he hand, neu ophilia and ele a ed se um c ea inine we e no associa ed wi h p olonged LOS in his analysis.7 Finally, a Canadian s udy based on he na ional da abase o 65,454 pa ien s admi ed wi h celluli is has shown ha o e 65 yea s o age, emale gende and conges i e hea ailu e we e ac o s associa ed wi h p olonged LOS.8 He e ogeneous esul s s ill exis and could be in lu- enced by he absence o clea ly de ined c i e ia o he di- agnosis and app oach o pa ien s wi h celluli is/e ysipelas, leading o di e en app oaches in he di e en cen es, as well as ega ding he a e age LOS.9–11 Howe e , Gang e al. a gued ha he esul s o hese s udies could ep esen an impo an ool o he cons uc ion o a se e i y sco e,4 in line wi h he sco es applied o pneumonia ha led o he educ ion o he a es o hospi al admission, as well as LOS and associa ed complica ions.12–15 The p esence o po en ial con ounding ac o s o p o- longed LOS is one o he majo limi a ions o his obse - a ional s udy, including un a ou able social backg ound o o he complica ions no di ec ly ela ed o he episode o celluli is/e ysipelas. Howe e , he applica ion o an im- p o ed logis ic eg ession model, wi h a s epwise da a en- y me hod, was aimed a educing he s a is ical impac o con ounding ac o s. CONCLUSION In conclusion, he knowledge o he p edic i e clinical and labo a o ial cha ac e is ics could be ele an o im- p o ed heal hca e deli e y, aimed a educing he leng h o s ay and associa ed isks and cos s. HUMAN AND ANIMAL PROTECTION The au ho s decla e ha he ollowed p ocedu es we e acco ding o egula ions es ablished by he E hics and Table 3 – Imp o ed logis ic eg ession, including 70% o da a (262/372), due o missing alues p- alue Adjus ed odds a io 95% con idence in e al Age 0.002 1.03 1.01 – 1.04 P e ious hospi al admission due o celluli is/ e ysipelas 0.005 4.81 1.63 –14.23 Pe iphe al ascula disease 0.42 1.30 0.69 – 2.46 Complica ions 0.001 3.28 2.63 – 6.59 Leukocy osis 0.049 1.81 1.00 – 3.30 CRP 0.035 1.03 1.00 – 1.06 Posi i e cul u e* 0.002 2.59 1.41 – 4.79 Roda A, e al. Fac o s associa ed wi h p olonged hospi al s ay among pa ien s wi h celluli is/e ysipelas, Ac a Med Po 2019 Jun;32(6):448–452 ARTIGO ORIGINAL Re is a Cien í ica da O dem dos Médicos www.ac amedicapo uguesa.com 452 Clinical Resea ch Commi ee and acco ding o he Helsinki Decla a ion o he Wo ld Medical Associa ion. DATA CONFIDENTIALITY The au ho s decla e ha hey ha e ollowed he p o o- cols o hei wo k cen e on he publica ion o pa ien da a. In o med consen s we e ob ained. CONFLICTS OF INTEREST The au ho s decla e ha he e we e no con lic s o in e - es in w i ing his manusc ip . FINANCIAL SUPPORT The au ho s decla e ha he e was no public o p i a e inancial suppo in w i ing his manusc ip . Roda A, e al. Fac o s associa ed wi h p olonged hospi al s ay among pa ien s wi h celluli is/e ysipelas, Ac a Med Po 2019 Jun;32(6):448–452 REFERENCES 1. Cae ano M, Amo im I. E ysipelas. Ac a Med Po . 2005;18:385–93. 2. S e ens DL, Bisno AL, Chambe s HF, Dellinge EP, Golds ein EJ, Go bach SL, e al. P ac ice guidelines o he diagnosis and managemen o skin and so issue in ec ions: 2014 upda e by he In ec ious Diseases Socie y o Ame ica. Clin In ec Dis. 2014;59:e10–52. 3. E on LJ, Lipsky BA, Low DE, Na hwani D, Tice AD, Vol u o GA. Expe panel on managing skin and so issue in ec ions. Managing skin and so issue in ec ions: expe panel ecommenda ions on key decision poin s. J An imic ob Chemo he . 2003;52:i3–17. 4. Ga g A, La ian J, Lin G, Sison C, Oppenheim M, Koo B. Clinical cha ac e is ics associa ed wi h days o discha ge among pa ien s admi ed wi h a p ima y diagnosis o lowe limb celluli is. J Am Acad De ma ol. 2017;76:626–31. 5. Mo pe h SC, Chambe s ST, Gallaghe K, F amp on C, Pi hie AD. Lowe limb celluli is: ea u es associa ed wi h leng h o hospi al s ay. J In ec . 2006;52:23–9. 6. Ca a alà J, Rosón B, Fe nández-Sabé N, Shaw E, del Rio O, Ri e a A, e al. Fac o s associa ed wi h complica ions and mo ali y in adul pa ien s hospi alized o in ec ious celluli is. Eu J Clin Mic obiol In ec Dis. 2003;22:151–7. 7. Fig ee M, Konecny P, Jennings Z, Goh C, K ilis SA, Miyakis S. Risk s a i ica ion and ou come o celluli is admi ed o hospi al. J In ec . 2010;60:431–9. 8. Baibe geno a A, D ucke AM, Shea NH. Hospi aliza ions o celluli is in Canada: a da abase s udy. J Cu an Med Su g. 2014;18:33–7. 9. Le ell NJ, Wing ield CG, Ga ioch JJ. Se e e lowe limb celluli is is bes diagnosed by de ma ologis s and managed wi h sha ed ca e be ween p ima y and seconda y ca e. B J De ma ol. 2011;164:1326–8. 10. A akaki RY, S azzula L, Woo E, K oshinsky D. The impac o de ma ology consul a ion on diagnos ic accu acy and an i- bio ic use among pa ien s wi h suspec ed celluli is seen a ou pa ien in e nal medicine o ices: a andomized clinical ial. JAMA De ma ol. 2014;150:1056–61. 11. S azzula L, Co lia J, Fox LP, Hughey L, Shinkai K, Gee SN, e al. Inpa ien de ma ology consul a ion aids diagnosis o celluli is among hospi alized pa ien s: a mul i-ins i u ional analysis. J Am Acad De ma ol. 2015;73:70–5. 12. Dean NC, Sil e MP, Ba eman KA, James B, Hadlock CJ, Hale D. Dec eased mo ali y a e implemen a ion o a ea men guideline o communi y-acqui ed pneumonia. Am J Med. 2001;110:451–7. 13. Dean NC, Suchy a MR, Ba eman KA, A onsky D, Hadlock CJ. Implemen a ion o admission decision suppo o communi y-acqui ed pneumonia: a pilo s udy. Ches . 2000;117:1368–77. 14. Menéndez R, To es A, Zalacaín R, Aspa J, Ma ín-Villascla as JJ, Bo de ías L, e al. Guidelines o he ea men o communi y-acqui ed pneumonia: p edic o s o adhe ence and ou come. Am J Respi C i Ca e. 2005;172:757–62. 15. Capelas egui A, Espana PP, Quin ana JM, A ei io I, Go o do I, Egu ola M, e al. Valida ion o a p edic i e ule o he managemen o communi y- acqui ed pneumonia. Eu Respi J. 2006;27:151–7.