- Journal List
- JAMA Network
- PMC5817617
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
JAMA Dermatology
JAMA Dermatol. 2017 Jun; 153(6): 578–582.
Published online 2017 Mar 15. doi:10.1001/jamadermatol.2017.0159
PMCID: PMC5817617
PMID: 28296993
Duncan R. Cranendonk, MD,
1,2,3 Lonneke A. van Vught, MD, PhD,1,2 Maryse A. Wiewel, MD, PhD,1,2 Olaf L. Cremer, MD, PhD,4 Janneke Horn, MD, PhD,5 Marc J. Bonten, MD, PhD,6 Marcus J. Schultz, MD, PhD,5 Tom van der Poll, MD, PhD,1,2,3 and W. Joost Wiersinga, MD, PhD1,2,3
Author information Article notes Copyright and License information PMC Disclaimer
Associated Data
- Supplementary Materials
This cohort study describes the clinical presentation and outcomes of patients with intensive care unit–necessitating cellulitis and compares them with patients with necrotizing fasciitis.
Abstract
Importance
Cellulitis is a commonly occurring skin and soft tissue infection and one of the most frequently seen dermatological diseases in the intensive care unit (ICU). However, clinical characteristics of patients with cellulitis requiring intensive care treatment are poorly defined. Necrotizing fasciitis is often confused for cellulitis at initial presentation and is considered to be more severe and thus has previously been described in more detail.
Objective
To describe the clinical presentation and outcomes of patients with ICU-necessitating cellulitis and to compare them with patients with necrotizing fasciitis.
Design, Setting, and Participants
This prospective cohort study includes all ICU admissions from 2 tertiary hospitals in the Netherlands. Of 2562 sepsis admissions, 101 had possible, probable, or definite cellulitis or soft tissue infections. Retrospective review identified severe cellulitis was the reason for ICU admission in 23 patients, necrotizing fasciitis in 31 patients, and other diagnoses in 47 patients.
Main Outcomes and Measures
Patient and disease characteristics, cultured pathogens, lengths of stay, and short-term and long-term mortality.
Results
Overall, 54 patients with cellulitis (n = 23; mean [SD] age, 57.2 [17.7] years) or necrotizing fasciitis (n = 31; mean [SD] age, 54.3 [13.5]) were included in this study. Patients with cellulitis were found to be less severely ill than patients with necrotizing fasciitis. This is reflected in rates of shock (7 [30.4%] vs 19 [61.3%]; P = .03), need for mechanical ventilation (12 [52.2%] vs 19 [93.5%]; P = .003) and slightly lower mean Sequential Organ Failure Assessment scores (8 vs 10; P = .046). Median (interquartile range [IQR]) Acute Physiology and Chronic Health Evaluation IV scores did not differ significantly (82 [75-98] vs 76 [70-96]; P = .16). Patients with cellulitis had more chronic comorbidities than patients with necrotizing fasciitis (20 [87.0%] vs 17 [54.8%]; P = .02), especially cardiovascular insufficiencies (10 [43.5%] vs 4 [12.9%]; P = .02) and immunodeficiencies (9 [39.1%] vs 3 [9.7%]; P = .02). Among patients with cellulitis and patients with patients with necrotizing fasciitis, Staphylococcus aureus (10 [43.5%] vs 4 [12.9%]; P = .02), Streptococcus pyogenes (2 [8.7%] vs 19 [61.3%]; P < .001) and Escherichia coli (4 [17.4%] vs 5 [16.2%]; P = .90) were the most frequently observed pathogens. Median (IQR) length of ICU stay was shorter for patients with cellulitis vs patients with necrotizing fasciitis (3 [2-5] vs 5 [3-11]; P = .01), while median (IQR) hospital length of stay did not differ significantly (22 [10.25-32] vs 36 [14.25-40]; P = .16); and the in-hospital mortality rate (26.1% vs 22.6%, P > .99) and 90-day mortality rate (30.4% vs 22.6%; P = .54) were similar.
Conclusions and Relevance
Patients with cellulitis patients are seldom admitted to the ICU. However, while these patients are less critically ill on admission than patients with necrotizing fasciitis, they have more chronic comorbidities and most notably similar short-term and long-term mortality.
Trial Registration
clinicaltrials.gov Identifier: NCT01905033
Key Points
Question
How do the clinical presentation and outcomes of patients with cellulitis who require intensive care compare with those of patients with necrotizing fasciitis?
Findings
In this cohort study of 2562 sepsis admissions to intensive care units at 2 tertiary hospitals in the Netherlands, 23 patients admitted for cellulitis and 31 patients admitted for necrotizing fasciitis were included in this study. Fewer patients with cellulitis had shock and fewer needed mechanical ventilation, but patients with cellulitis had more chronic comorbidities; in-hospital and 90-day mortality were similar.
Meaning
Physicians should be aware that patients admitted to the intensive care unit for cellulitis and necrotizing fasciitis have similar mortality risks, despite differences in initial presentation.
Introduction
Skin and soft tissue infections, in particular cellulitis, are among the most common infections leading to hospitalization. Skin and soft tissue infections account for 7% of infections in patients admitted for sepsis in the intensive care unit (ICU). The 2 most prevalent skin and soft tissue infections in the ICU setting are cellulitis and necrotizing fasciitis (NF). While there are numerous reports on critically ill patients with NF, little is known about patients with severe (here defined as ICU-necessitating) cellulitis, who account for less than 2.5% of all patients hospitalized for cellulitis. By examining a large prospective ICU cohort, our aim was to provide insight into the clinical presentation and outcome of patients admitted to the ICU because of severe cellulitis. At first presentation, severe cellulitis and NF are often indistinguishable. Therefore, as a comparator, we also examined all patients admitted to the ICU with NF.
Methods
This was a substudy of the “Molecular Diagnosis and Risk Stratification of Sepsis” (MARS) project, a prospective observational study in 2 tertiary ICUs. All adult patients expected to stay more than 24 hours admitted between January 2011 and January 2014 were included via an opt-out method approved by the institutional review boards of both hospitals; participants were informed about the study by a brochure provided at ICU admission with an opt-out card that could be completed by the patient or a legal representative in case of unwillingness to participate. For every patient the plausibility of an infection was assessed (ie, “none,” “possible,” “probable,” and “definite”) using consensus definitions as described (eMethods and eTable 1 in the Supplement). Patients with a possible, probable, or definite cellulitis or soft tissue infection were selected, and an independent researcher performed a retrospective medical record review to determine an exact diagnosis and primary reason for each included ICU admission (eMethods in the Supplement). Thus, patients were marked as admissions for severe cellulitis, NF, another diagnosis but with cellulitis as a comorbidity, or excluded because they had neither cellulitis nor NF. Detailed methods are available in the eMethods in the Supplement.
Results
Of 2562 sepsis admissions, 101 with cellulitis or soft tissue infection as primary or secondary diagnosis met our selection criteria (3.9%); 26 with possible, 13 with probable and 62 with definite likelihood. Retrospective medical review of these patients led to the identification of 23 patients with severe cellulitis (0.9%; 6 initially scored as possible; 3, probable; and 14, definite) and 31 patients with NF (1.2%; 1 initially scored as possible; 30, definite). Twenty-five patients were admitted with another primary diagnosis but had cellulitis as comorbidity and were not analyzed further. Twenty-two patients did not meet the criteria for cellulitis or NF and were excluded (eTable 2 in the Supplement).
Baseline characteristics between patients with severe cellulitis and NF were comparable in terms of age and race (Table 1). No statistically significant difference was seen in body mass index (32.4 for cellulitis vs 27.3 for NF [calculated as weight in kilograms divided by height in meters squared]; P = .06). Compared with NF, patients with severe cellulitis were more often medical admissions (65.2% vs 29.0%; P = .01) and had more chronic comorbidities (87% vs 54.8%; P = .02), especially cardiovascular insufficiencies (43.5% vs 12.9%; P = .02) and immunodeficiencies (39.1% vs 9.7%; P = .02).
Table 1.
Demographics and Clinical Characteristics of 54 Patients
| Characteristic | Cellulitis (n = 23) | Necrotizing Fasciitis (n = 31) | P Value |
|---|---|---|---|
| Age, mean (SD) | 57.2 (17.7) | 54.3 (13.5) | .51 |
| Race, white, No. (%) | 19 (83) | 29 (94) | .35 |
| Male, No. (%) | 10 (43) | 20 (65) | .18 |
| BMI, mean (SD) | 32.4 (11.1) | 27.3 (5.7) | .06 |
| Medical admission, No. (%) | 15 (65) | 9 (29) | .01 |
| Admission source, No. (%) | .09 | ||
| Emergency department | 11 (48) | 11 (35) | |
| ICU other hospital | 2 (9) | 12 (39) | |
| Nursing ward | 4 (17) | 1 (3) | |
| Operating theater | 5 (22) | 5 (16) | |
| Coronary or medium care facility | 1 (4) | 2 (6) | |
| Days in hospital prior to ICU admission, median (IQR) | 1 (0-3) | 1 (1-3) | .18 |
| Shock, No. (%) | 7 (30) | 19 (61) | .03 |
| Mechanical ventilation, No. (%) | 12 (52) | 29 (94) | .003 |
| Max temperature first 24 h, median (IQR) | 37.9 (36.6-39.0) | 37.9 (36.9-38.8) | .92 |
| APACHE IV, median (IQR) | |||
| Score | 82 (75-98) | 76 (70-96) | .16 |
| APS | 72 (62.5-78.5) | 71 (53.0-84.0) | .64 |
| PMR | 0.28 (0.18-0.56) | 0.22 (0.13-0.4) | .17 |
| SOFA score, median (IQR) | 8 (6-10) | 10 (8-12) | .046 |
| Organ failure, No. (%) | 21 (91) | 27 (87) | >.99 |
| Concurrent infection (%) | 6 (26) | 4 (13) | .30 |
| Chronic comorbidity, No. (%) | 20 (87) | 17 (55) | .02 |
| Cardiovascular insufficiencies | 10 (43) | 4 (13) | .02 |
| Diabetes mellitus | 9 (39) | 7 (23) | .23 |
| Immunocompromised | 9 (39) | 3 (10) | .02 |
| HIV | 2 (9) | 0 | .19 |
| Renal insufficiency | 6 (26) | 2 (6) | .06 |
| Charlson comorbidity index, median (IQR) | 4 (3-5) | 3 (1-4) | .17 |
| Malignancy | 2 (9) | 6 (19) | .47 |
| Respiratory insufficiency | 3 (13) | 4 (13) | >.99 |
| Site of infection, No. (%) | <.001 | ||
| Head/neck | 0 | 4 (13) | |
| Torso | 2 (9) | 7 (23) | |
| Upper extremities | 1 (4) | 8 (26) | |
| Lower extremities | 14 (61) | 3 (10) | |
| Genital/perineal | 0 | 6 (19) | |
| Sacrum | 2 (9) | 0 | |
| Unknown | 1 (4) | 0 | |
| Multiple locationsa | 3 (13) | 3 (10) | |
| First CRP during first 24 h, median (IQR), mg/L | 211 (55-288) | 223 (167-309) | .38 |
| Peak, median (IQR) | |||
| Leukocytes during first 24 h, ×109/L | 12.6 (7.7-17.5) | 15.7 (12.2-22.2) | .31 |
| Lactate during first 24 h, mmol/L | 2.4 (2.0-3.4) | 3.3 (2.2-6.4) | .18 |
| Creatinine during first 24 h, µmol/L | 169 (123-253) | 178 (128-249) | .97 |
Open in a separate window
Abbreviations: APACHE, acute physiology and chronic health evaluation; APS, acute physiology score; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CRP, C-reactive protein; HIV, human immunodeficiency virus; ICU, intensive care unit; IQR, interquartile range; PMR, predicted mortality rate; SOFA, sequential organ failure assessment.
aMultiple locations included lower extremity and torso (n = 1), upper extremity and torso (n = 1), upper and lower extremity (n = 1) in severe cellulitis; and lower extremity and torso (n = 2), upper extremity and torso (n = 1) in necrotizing fasciitis.
On ICU admission, patients with severe cellulitis had mean APACHE IV scores comparable to patients with NF, but their mean sequential organ failure assessment (SOFA) scores were slightly lower (8 vs 10; P = .046), and fewer had shock (30.4% vs 61.3%; P = .03) or need for mechanical ventilation (52.2% vs 93.5%; P = .003). Most severe cellulitis cases involved the lower extremities while NF presented mainly in the upper body. Serum C-reactive protein, leukocyte, creatinine and lactate levels were comparable between groups.
Six patients with severe cellulitis and all patients with NF underwent surgery, resulting in 1 and 2 amputations, respectively. The amputation for the patient with severe cellulitis was due to arterial occlusion, not infection. Other patients with severe cellulitis surgery indications were for ruling out NF (3 patients), cutaneous fistula closure (1 patient), and abscess drainage (1 patient). Two patients with severe cellulitis (8.7%) and 15 patients with NF (48.4%) received at least 1 dose of intravenous immunoglobulins.
No significant differences were observed regarding ICU-acquired complications (acute myocardial, kidney or lung injury, ICU-acquired weakness, or pneumothorax), occurring in 13% of patients with severe cellulitis vs 22.6% of patients with NF (P = .47). Length of ICU stay was shorter for patients with severe cellulitis (Table 2). Remarkably, no significant differences were seen for ICU mortality, in-hospital mortality, and 30-day and 90-day mortality. In-hospital observed mortality is consistent with APACHE IV predictions, with standardized mortality ratios for severe cellulitis (0.93) and NF (1.03) approximating 1.0.
Table 2.
Causative Pathogens and Outcomes for 54 Patients
| Causative Pathogens | Cellulitis (n = 23) | Necrotizing Fasciitis (n = 31) | P Value |
|---|---|---|---|
| Gram-positive bacteria, No. (%) | 16 (70) | 30 (97) | .01 |
| Clostridium species | 0 | 2 (6) | .22 |
| Enterococcus | |||
| faecalis | 1 (4) | 0 | .25 |
| species | 1 (4) | 1 (3) | .83 |
| Staphylococcus aureus | 10 (43) | 4 (13) | .02 |
| Streptococcus | |||
| agalactiae | 0 | 1 (3) | .39 |
| pyogenes | 2 (9) | 19 (61) | <.001 |
| species | 1 (4) | 3 (10) | .47 |
| viridans | 1 (4) | 0 | .25 |
| Gram-negative bacteria, No. (%) | 10 (43) | 10 (32) | .41 |
| Bacteroides species | 1 (4) | 2 (6) | .74 |
| Enterobacter cloacae | 1 (4) | 2 (6) | .74 |
| Escherichia coli | 4 (17) | 5 (16) | .90 |
| Proteus mirabilis | 1 (4) | 0 | .25 |
| Pseudomonas aeruginosa | 3 (13) | 1 (3) | .22 |
| Type of infection, No. (%) | .14 | ||
| Monomicrobial infection | 12 (52) | 21 (68) | |
| Polymicrobial infection | 6 (26) | 9 (29) | |
| Unknown, no pathogens cultured | 5 (22) | 1 (3) | |
| Necrotizing fasciitis | |||
| Type I | 0 | 10 (32) | |
| Type II | 0 | 18 (58) | |
| Uncertain/other type | 0 | 3 (10) | |
| Outcomes, median (IQR), d | |||
| ICU length of stay | 3 (2-5) | 5 (3-11) | .01 |
| Hospital length of stay | 22 (10.25-34.00) | 36 (14.25-40.00) | .16 |
| Mortality, No. (%) | |||
| ICU | 2 (9) | 6 (19) | .45 |
| Hospital | 6 (26) | 7 (23) | >.99 |
| 30-d | 5 (22) | 7 (23) | >.99 |
| 90-d | 7 (30) | 7 (23) | .54 |
Open in a separate window
Abbreviations: ICU, intensive care unit; IQR, interquartile range.
A sensitivity analysis, excluding patients with possible likelihoods, shows similar results (eTable 3 in the Supplement).
Discussion
Our objective was to describe the clinical characteristics and outcomes of patients admitted to the ICU for severe cellulitis in comparison with patients admitted for NF. Using a unique, well-documented prospective cohort of critically ill patients with sepsis, we found that patients with severe cellulitis have similar in-hospital and long-term mortality rates when compared with patients admitted for NF. Compared with NF, severe cellulitis patients were less severely ill on ICU admission as reflected by a lower occurrence of shock, need for mechanical ventilation, and presence of organ failure, despite more chronic comorbidities.
The in-hospital mortality rate of the patients with severe cellulitis and NF included in our study was 26% and 23%, respectively. George et al also reported similar, albeit higher, mortalities, of 43% and 42% for severe cellulitis and NF. They attributed their high mortality to selection of the severest cases due to scarcity of critical care resources in the United Kingdom. One review of 67 studies found an overall NF mortality of 23.5%. In another report describing ICU admissions due to suspected NF, 12 patients were eventually diagnosed with cellulitis, none of whom died. However, none of those 12 patients were in shock, and only 2 required mechanical ventilation, suggesting that they might have been less ill than the patients in our study.
While patients with severe cellulitis at presentation were less sick in comparison with NF patients, both their in-hospital and long-term mortality were similar. Our data suggest this discrepancy may be explained by a higher frequency of chronic comorbidities in patients with severe cellulitis. This is in line with previous studies on the subject.
To our knowledge, this study is the first to describe detailed clinical characteristics and outcomes of patients with severe cellulitis and provides a context by comparing those clinical characteristics with those of patients with NF. One of the strengths of our study is the prospective data collection and retrospective in-depth diagnostic review, improving the accuracy of severe cellulitis and NF as primary diagnosis for ICU admission and lowering the risk of information bias. There is also a low selection bias risk as all consecutively admitted patients were enrolled. Municipal mortality data ensured a reliable reporting of long-term mortality.
Limitations
This study has limitations. The number of patients with severe cellulitis and NF in this large cohort was limited. Because multiple variables were analyzed, we are subject to the effects of multiple testing and thus potential type I errors. Additionally, all patients were admitted to tertiary ICUs, so extrapolation to a general ICU population should be done with caution. Misdiagnosis of cellulitis can occur in 30% of patients hospitalized for cellulitis. The patients with severe cellulitis in our study were unlikely to suffer from misdiagnosis, as so-called pseudocellulitis diagnoses rarely cause critical illness, but it cannot be ruled out. Finally, these results should not be interpreted as cellulitis being equally debilitating as NF.
Conclusions
In conclusion, in a cohort of over 2500 well-documented patients with sepsis admitted to the ICU, 2.1% of patients had severe cellulitis or NF as their primary diagnosis. Although patients with severe cellulitis were less severely ill on admission, both in-hospital and long-term mortality was similar between patients with severe cellulitis and NF.
Notes
Supplement.
eAppendix 1. Full list of members of the MARS consortium
eAppendix 2. Detailed study methods
eTable 1. Diagnostic criteria for cellulitis and soft tissue infections, adapted from Klein Klouwenburg et al
eTable 2. Skin or soft tissue findings on admission of patients scored as possible, probable or definite cellulitis or soft tissue infection, but who were not admitted for severe cellulitis or necrotizing fasciitis
eTable 3. Sensitivity analysis excluding patients with a possible likelihood of cellulitis or soft tissue infection.
eReferences.
Click here for additional data file.(63K, pdf)
References
1. Raff AB, Kroshinsky D. Cellulitis: a review. JAMA. 2016;316(3):325-337. [PubMed] [Google Scholar]
2. Christensen KL, Holman RC, Steiner CA, Sejvar JJ, Stoll BJ, Schonberger LB. Infectious disease hospitalizations in the United States. Clin Infect Dis. 2009;49(7):1025-1035. [PubMed] [Google Scholar]
3. Vincent JL, Rello J, Marshall J, et al.; EPIC II Group of Investigators . International study of the prevalence and outcomes of infection in intensive care units. JAMA. 2009;302(21):2323-2329. [PubMed] [Google Scholar]
4. George SM, Harrison DA, Welch CA, Nolan KM, Friedmann PS. Dermatological conditions in intensive care: a secondary analysis of the Intensive Care National Audit and Research Centre (ICNARC) Case Mix Programme database. Crit Care. 2008;12(suppl 1):S1. [PMC free article] [PubMed] [Google Scholar]
5. Kennedy M, Joyce N, Howell MD, Lawrence Mottley J, Shapiro NI. Identifying infected emergency department patients admitted to the hospital ward at risk of clinical deterioration and intensive care unit transfer. Acad Emerg Med. 2010;17(10):1080-1085. [PMC free article] [PubMed] [Google Scholar]
6. van Vught LA, Klein Klouwenberg PM, Spitoni C, et al.; MARS Consortium . Incidence, risk factors, and attributable mortality of secondary infections in the intensive care unit after admission for sepsis. JAMA. 2016;315(14):1469-1479. [PubMed] [Google Scholar]
7. May AK.Skin and soft tissue infections. Surg Clin North Am. 2009;89(2):403-420, viii. [PubMed] [Google Scholar]
8. Zahar JR, Goveia J, Lesprit P, Brun-Buisson C. Severe soft tissue infections of the extremities in patients admitted to an intensive care unit. Clin Microbiol Infect. 2005;11(1):79-82. [PubMed] [Google Scholar]
9. Zimmerman JE, Kramer AA, McNair DS, Malila FM. Acute Physiology and Chronic Health Evaluation (APACHE) IV: hospital mortality assessment for today’s critically ill patients. Crit Care Med. 2006;34(5):1297-1310. [PubMed] [Google Scholar]
10. Davis JS, He V, Anstey NM, Condon JR. Long term outcomes following hospital admission for sepsis using relative survival analysis: a prospective cohort study of 1,092 patients with 5 year follow up. PLoS One. 2014;9(12):e112224. [PMC free article] [PubMed] [Google Scholar]
11. Lemay AC, Anzueto A, Restrepo MI, Mortensen EM. Predictors of long-term mortality after severe sepsis in the elderly. Am J Med Sci. 2014;347(4):282-288. [PMC free article] [PubMed] [Google Scholar]
12. Weng QY, Raff AB, Cohen JM, et al.. Costs and consequences associated with misdiagnosed lower extremity cellulitis [published online November 2, 2016]. JAMA Dermatol. doi: 10.1001/jamadermatol.2016.3816 [PubMed] [CrossRef] [Google Scholar]
13. Obaitan I, Dwyer R, Lipworth AD, et al.. Failure of antibiotics in cellulitis trials: a systematic review and meta-analysis. Am J Emerg Med. 2016;34(8):1645-1652. [PubMed] [Google Scholar]
Articles from JAMA Dermatology are provided here courtesy of American Medical Association
