Ultrasound

Ultrasound for the diagnosis of necrotizing fasciitis: A systematic review of the literature

a b s t r a c t

Background: necrotizing fasciitis (NF) is a deadly disorder that can be challenging to diagnose on history and examination alone. Point-of-care ultrasound is widely available and has been increasingly used for diagnosing skin and soft tissue infections. We performed a systematic review to determine the accuracy of POCUS for diagnosing NF with subgroup analyses of the accuracy of specific POCUS examination components. Methods: PubMed, Scopus, CINAHL, LILACS, the Cochrane databases, Google Scholar, and bibliographies of selected articles were assessed for all retrospective, prospective, and randomized control trials evaluating the accuracy of POCUS for diagnosing NF. Data were dual extracted into a predefined worksheet and quality analysis was performed with the QUADAS-2 tool. Data were summarized and an overall summary was completed.

Results: We identified three papers (n = 221 patients; 33% NF) that met our inclusion criteria. The overall sensitivity ranged from 85.4%-100% while the specificity ranged from 44.7% to 98.2%. fluid accumulation along the fascial plane was the most sensitive (85.4%; 95% CI 72.2% – 93.9%), while subcutaneous emphysema was the most specific (100%; 95% CI 92.5% – 100%).

Conclusions: POCUS has good sensitivity and specificity for the diagnosis of NF. POCUS should be considered as an adjunct to the initial clinical decision making for the diagnosis of NF.

(C) 2022

  1. Introduction

Necrotizing Fasciitis (NF) is an emergent life-threatening condition that requires early diagnosis in order to be appropriately treated [1]. NF is a subset of soft tissue infections that is characterized by rapidly spreading tissue destruction, fulminant sepsis, and high mortality [1,2]. Mortality from NF increases dramatically when there is a Delay in diagnosis and subsequent treatment, which includes rapid mobiliza- tion of emergency department resources including broad spectrum antibiotics, aggressive supportive care, and prompt surgical interven- tion [3-5]. Early in the course of NF, overlying tissue can appear normal or consistent with mild skin inflammation, which can make Timely diagnosis challenging [6]. By the time skin findings more typical of NF (e.g., erythema, bullae) have emerged, the patient is more severely ill and the mortality rate is significantly higher [3-5]. Therefore, there is a need to better identify potential NF cases more quickly in the ED.

In response to this, tools such as the Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) score have been used in conjunction

* Corresponding author at: 1750 West Harrison Street, Suite 108 Kellogg, Chicago, IL 60612, United States of America.

E-mail address: [email protected] (A. Marks).

with history and examination to aid in diagnosis. However, studies have suggested that they are insufficiently sensitive to diagnose NF [7,8]. One systematic review examined history, physical examination, and the LRINEC score and found low sensitivity for all of these diagnostic modalities [9].

Point-of-care ultrasound is a diagnostic modality readily available in most emergency departments and has been demonstrated to have superior diagnostic performance for skin and soft tissue abscesses compared with physical examination alone [10]. Building upon this, POCUS has also been proposed as a tool to identify NF early in the course of the disease with several more recent studies assessing this application. While computed tomography (CT) imaging is useful in identifying NF, it exposes patients to radiation and requires the patient to leave the Emergency Department (ED). POCUS in combina- tion with clinical examination and pre-test probability can assist in a more rapid diagnosis and potentially lead to a prompt surgical consulta- tion or advanced imaging. Additionally, POCUS can allow for more rapid reassessments in the case of changes in clinical condition. However, there is a need to better understand the diagnostic accuracy of this modality.

The primary objective of this study was to determine the accuracy of POCUS for diagnosing NF. We also planned a priori subgroup analyses of specific POCUS examination components for diagnosing NF.

https://doi.org/10.1016/j.ajem.2022.12.037

0735-6757/(C) 2022

  1. Methods

We conducted this study in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses – Diagnostic test accuracy (PRISMA-DTA) guidelines for systematic reviews using best practice guidelines [11]. This review was registered with PROSPERO (CRD42022324135). Utilizing a medical librarian, we conducted a search of PubMed, Scopus, the Cumulative Index of Nursing and Allied Health (CINAHL), the Latin American and Caribbean Health Sciences Literature database (LILACS), Google Scholar, the Cochrane Database of Systematic Reviews, and the Cochrane Central Register of Controlled Trials to include citations from inception to April 7th, 2022. Details of the search strategy are included in the Appendix. We reviewed the bibliographies of all included studies and review articles for potential missed articles. We also consulted with topic experts to help identify any further relevant studies. We reviewed clinicaltrials.gov to assess for any ongoing studies.

    1. Inclusion and exclusion criteria

Inclusion criteria consisted of all retrospective or prospective studies evaluating ultrasound for the diagnosis of necrotizing fasciitis. There were no language, date, or age restrictions. We excluded case reports, case series, and review articles. Two investigators independently assessed studies for eligibility based upon the criteria. All titles and abstracts meeting initial criteria were reviewed as full-text manuscripts. Studies determined to meet the eligibility criteria on full-text review by both extractors were included in the final data analysis. Any discrepan- cies were resolved by consensus with the addition of a third reviewer if needed.

    1. Data collection and processing

Two investigators independently extracted data from the included studies. The investigators underwent initial training and extracted data into a pre-designed, piloted data collection form. The following in- formation was abstracted: last name of the first author, Publication year, study country, study population size, type of study (e.g., retrospective, prospective observational, randomized controlled trial), study location (eg, ED, Intensive Care Unit), mean age of study patients, sex of the study patients, number of cases of necrotizing fasciitis, ultrasound trans- ducer type (e.g., linear, curvilinear), Ultrasound technique, diagnostic criteria, sonographer training, sonographer specialty, sonographer experience (e.g., attending physician, resident physician), and true positives, false positives, true negatives, and false negatives.

Studies were independently assessed for quality by two separate investigators utilizing the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool [12]. Any discrepancies were resolved by consensus with the addition of a third reviewer if needed. The authors evaluated risk of bias and degree of applicability for each individual study using the following criteria:

      • In the assessment of “patient selection,” studies that included consecutively recruited patients or a random selection of patients were considered “low risk” of bias. convenience sampling was considered “high risk” of bias. If the recruitment was not described, studies were considered at “unclear risk” of bias. As the study question was intended to assess POCUS, all examinations performed by a non-radiologist clinician at the bedside were considered “low risk” for applicability. Studies performed by a radiologist or radiology technician were considered “high risk” for applicability. If the sonog- rapher was not described, studies were considered at “unclear risk” of applicability.
      • In the assessment of “index test,” studies were determined to be “low risk” of bias if the sonographer was blinded to the results of the reference standard. If the sonographer was not blinded, the study

was deemed “high risk” of bias. If blinding was not described, studies were considered at “unclear risk” of bias. With regard to applicability, studies were deemed “low risk” if they used a discrete ultrasound protocol. Studies that did not utilize an ultrasound protocol were deemed “high risk” of applicability. If the protocol was not described, studies were considered at “unclear risk” of applicability.

      • In the assessment of “reference standard,” studies were determined to be “low risk” of bias if the reference standard (e.g., CT, operative diagnosis) was made blinded to the results of the index test. If the reference standard diagnosis was not blinded to the POCUS, the study was deemed “high risk” of bias. If blinding was not described, studies were considered at “unclear risk” of bias. With regard to applicability, studies were deemed “low risk” if they used a single, acceptable gold standard (e.g., CT, operative diagnosis). Studies were deemed at “high risk” for applicability if they used multiple different reference standards.
      • In the assessment of “flow and timing,” we defined “low risk” studies as those where all participants were followed up and had the same reference standard. Studies with different reference standards for some patients were deemed “unclear risk” of bias. Studies where more than half of patients did not receive a reference standard or were lost to follow up were deemed “high risk” of bias.
    1. Statistical analyses

We created two-by-two contingency tables for each study with sensitivities and specificities and 95% confidence intervals (CIs) derived from this data. Data were not combined for meta-analysis due to the small number of studies and significant clinical heterogeneity with regard to the diagnostic criteria.

  1. Results

We identified 867 total studies. PubMed identified 323 studies, Scopus retrieved 236 studies, CINAHL found 109 studies, the Cochrane Central Register of Controlled Trials located 99 studies, and both LILACS and the Cochrane Database of Systematic Reviews identified no studies. In addition, the initial 100 studies from Google Scholar were also included as recommended by Bramer and colleagues [13]. After removing duplicates, 649 original abstracts were reviewed with 23 selected for full text review and 3 identified for inclusion (Fig. 1). No additional papers were identified by the topic experts or through biblio- graphic review. No trials were identified through clinicaltrials.gov.

We identified three total papers (n = 221 patients) which met our inclusion criteria. Two studies were conducted in Taiwan [14,15] and one in the United States [16] (Table 1). All three were conducted in the ED. [14-16] Two were prospective [14,16], and one was retrospec- tive [15]. The mean age of participants was 57 years and 34% were female. Seventy-three of the 221 patients (33%) had necrotizing fasciitis. All ultrasound examinations were performed with a linear transducer [14-16]. All exams were performed by Emergency Medicine physicians, with two performed only by attending physicians [14,15], while the third included resident, fellow, or attending physicians [16].

All three studies reported diagnostic accuracy but the diagnostic criteria differed, so these were reported qualitatively and without meta-analysis [14-16]. The sensitivity ranged from 85.4% to 100%, while the specificity ranged from 44.7% to 98.2% (Table 3). One study

[15] reported accuracy stratified by individual ultrasound finding.

They found that fluid accumulation along the fascia was the most sensitive (85.4%; 95% CI 72.2% – 93.9%), while the most specific finding was subcutaneous emphysema (100%; 95% CI 92.5% – 100%) (Table 4). All three studies were at high risk of patient selection bias due to convenience sampling [14-16] (Table 2). All three studies were also at unclear risk of bias for both the index test and reference standard, as

Full-text articles excluded (n = 20)

Wrong study design (n = 19) Wrong outcome (n = 1)

Fig. 1. PRISMA flow diagram.

Records after duplicates removed (n = 649)

Additional records identified through other sources

(n = 0)

Records identified through database searching

(n = 867)

Full-text articles assessed for eligibility

(n = 23)

Records screened (n = 649)

Studies included in quantitative synthesis (meta-analysis)

(n = 0)

Studies included in qualitative synthesis (n = 3)

Records excluded (n = 626)

Included

they did not state whether the ultrasound was performed prior to the surgical consult or if the surgeon or pathologist were blinded to the ultrasound findings [14-16]. All three studies were at low risk of applicability concerns for patient selection and index test [14-16]. One study was at low risk of applicability concerns for reference standards [14], while two were at high risk because two different reference standards were used [15,16].

  1. Discussion

This systematic review found that POCUS has good sensitivity and specificity for the diagnosis of NF. To our knowledge, this is the first systematic review evaluating the diagnostic accuracy of POCUS for

NF. This review had several strengths, including a Comprehensive literature search of multiple databases with the assistance of an experienced medical librarian and identification of data on both overall sensitivity and specificity for POCUS, as well as that for specific exam findings.

While POCUS did not demonstrate perfect diagnostic accuracy, it is important to situate this within existing research on other tools for diagnosis. POCUS was significantly more sensitive (85.4% to 100%) than fever (46%), Hemorrhagic bullae (25%), or hypotension (21%), and rivaled CT (88.5%) [9]. It was also more sensitive than the LRINEC score even when using the lower threshold of >=6 [9]. In contrast, POCUS had more variable specificity, ranging from 44.7% to 98.2%, which was comparable to the above indicators (77% to 97%) [9].

Table 1

Characteristics of the included studies.

Study Study population

Country Study

Location

Mean patient age

Number (and %) female patients

Total cases of Necrotizing Fasciitis (%)

Ultrasound transducer

Ultrasound Diagnostic Criteria operator experience

Lanham 2021

64 United

States

ED 47 26 (40.6%) 8 (12.5%) Linear Presence of thickened fascial planes with fluid accumulation

in the facial layers, distortion of facial planes and presence of subcutaneous gas with dirty shadowing.

Resident, Fellow, Attendings

Lin

2019

Yen

2002

95 Taiwan ED 67 27 (28.4%) 48 (50.5%) Linear Ultrasonographic finding of fluid accumulation with a cutoff

point of >=2 mm of depth along the deep fascia

62 Taiwan ED 52.7 22 (35.5%) 17 (27.4%) Linear diffuse thickening of the subcutaneous tissue, accompanied

by a layer of fluid accumulation >4 mm in depth along the deep fascial layer when compared with the contralateral position on the corresponding normal limb

Attendings Attendings

Eligibility

Screening

Identification

ED, Emergency Department.

Table 2

QUADAS-2 for included studies.

Study

Risk of Bias

Applicability Concerns

Patient Selection

Index Test

Reference Standard

Flow and Timing

Patient Selection

Index Test

Reference Standard

Lanham 2021

H

U

U

U

L

L

H

Lin 2019

H

U

U

U

L

L

H

Yen 2002

H

U

U

U

L

L

L

L, low risk of bias; U, unclear risk of bias; H, high risk of bias.

Table 3

Diagnostic accuracy of ultrasound for necrotizing skin and soft tissue infections.

Study

Sensitivity (95% CI)

Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

Lahham 2021

100% (63.1% to 100%)

98.2% (90.5% to 100%%)

88.9% (53.4% to 98.2%)

100%

Yen 2002

88.2% (63.6% to 98.5%)

93.3% (81.7% to 98.6%)

83.3% (62.3% to 93.8%)

95.5% (85.1% to 98.7%)

Lin 2019

85.4% (72.2% to 93.9%)

44.7% (30.2% to 59.9%)

61.2% (54.3% to 67.7%)

75.0% (58.5% to 86.5%)

CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.

Table 4

Diagnostic accuracy of specific exam components.

Study?

Sensitivity (95% CI)

Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

Fluid accumulation

85.4% (72.2% to 93.9%)

44.7% (30.2% to 59.9%)

61.2% (54.3% to 67.7%)

75.0% (58.5% to 86.5%)

Irregular or thickened fascia

66.7% (51.6% to 79.6%)

55.3% (40.1% to 69.8%)

60.4% (51.1% to 68.9%)

61.9% (50.3% to 72.3%)

Subcutaneous cobblestoning

16.7% (7.5% to 30.2%)

74.5% (59.7% to 86.1%)

40.0% (23.1% to 59.7%)

46.7% (41.5% to 51.9%)

Subcutaneous emphysema

6.3% (1.3% to 17.2%)

100% (92.5% to 100%)

100%

51.1% (49.3% to 52.9%)

* Only Lin 2019 reported on accuracy of individual findings, so each are reported separately here; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.

It is worth noting several differences between studies, which may provide insight into the differences in sensitivity and specificity. Lin et al. was a retrospective study assessing the accuracy of multiple indi- vidual findings seen on POCUS for diagnosing necrotizing fasciitis [15]. However, the authors ultimately relied upon the presence of >=2 mm of fluid in the fascial plane as their primary criteria to diagnose necrotizing fasciitis and did not report the accuracy of combined findings. This dif- fers from Lahham et al. and Yen et al. [14,16], which used a combination of findings as part of their diagnostic criteria. Therefore, it is possible that combining several findings together may be more beneficial than relying upon a single finding in isolation and more data are needed to determine the ideal combination of findings need to diagnose necrotiz- ing fasciitis with the best combination of sensitivity and specificity. However, based upon the current data, the criteria by Lahham et al. appear to be most reasonable to use given the combination of both high sensitivity and high specificity.

One of the challenges in the included studies was variation in the diagnostic criteria used with POCUS. Traditionally, the diagnostic ultrasound findings consistent with NF can include fascial and subcutaneous tissue thickening, abnormal fluid accumulation in the deep fascia layer, and, in advanced cases, subcutaneous air [14,17]. One retrospective review of 32 pathologically-confirmed cases of NF showed that ultrasound revealed changes in subcutaneous fat (87.5%), underlying fascia (56%), and muscle (46.8%) [18]. Lin et al. specifically examined the sensitivity and specificity of each ultra- sound finding and found that the fluid accumulation and the occur- rence of an irregular or thickened fascial layer were significantly different between the NF and non-NF groups [15]. Lin et al. also examined the amount of fluid accumulation, ranging from 1 mm to 5 mm, and found that 2 mm of fluid had the best accuracy with a sensitivity of 75%, and a specificity of 70.2% [15].

To perform ultrasound for NF, a linear transducer is generally

utilized and placed directly on the area of concern to examine the skin, subcutaneous tissue, muscle and fascial layer to evaluate for the presence of air, thickening and fluid as discussed above. It is important to note that even one of the ultrasound findings of NF can aid in

diagnosis and not all of them must be present. There are also potential false negatives with this exam if focused on examination for all ultra- sound findings. For example, not all NF will be caused by gas-forming organisms and air may not be present on ultrasound. There are many disease processes that can mimic NF on ultrasound such as an open wound or a recently performed incision and drainage with tracked air. Thus, the entire clinical picture and laboratory values must be used in conjunction with the ultrasound findings [19].

Future studies should determine the ideal ultrasound diagnostic

criteria and externally validate the protocol across multiple users and setting. It would also be beneficial for future studies to integrate clinical exam findings or already validated scores for the diagnosis of NF, such as the LRINEC score, with POCUS findings to assess if it leads to improved diagnosis. Additionally, data are needed to determine the ideal timing of POCUS, the role of repeat or serial POCUS, and the impact of POCUS on clinical outcomes.

  1. Limitations

Several limitations must be considered in regard to the current re- view. One major limitation of this systematic review is the limited num- ber of studies evaluating POCUS for the diagnosis of NF. Only three studies that fit our inclusion criteria were identified and they did not use the same diagnostic criteria, which limited our ability to perform a meta-analysis for more inclusive data. Second, all three studies were performed with convenience samples. This may not reflect the full range of patients that would present to the emergency department. Third, it was unclear how blinding occurred in the studies, as there was no mention as to whether the pathologist/interpreter of the biop- sies was blinded to the ultrasound results. Most of the data collected in these studies were performed by ED attendings and it is unclear how the data would change with more novice users. Additionally, while we conducted a comprehensive search of multiple databases sup- plemented by expert consultation and bibliographic searching, it re- mains possible that studies may have been missed, though we believe this risk is low.

  1. Conclusion

POCUS is sensitive and specific for the diagnosis of NF and can be used as an adjunct to aid in Prompt diagnosis. Future studies are needed to determine the ideal protocol, effect of timing, role of repeat POCUS examinations, and impact on patient relevant outcomes.

Disclosures/Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. We have no disclosures to declare.

Meetings

None.

IRB

N/A.

CRediT authorship contribution statement

Amy Marks: Writing – review & editing, Writing – original draft, Formal analysis, Data curation, Conceptualization. Daven Patel: Formal analysis. Tina Sundaram: Formal analysis. Jordan Johnson: Writing – review & editing, Writing – original draft. Michael Gottlieb: Writing – review & editing, Writing – original draft, Supervision, Methodology, Formal analysis, Data curation, Conceptualization.

Acknowledgements

The authors thank Jennifer C. Westrick, MSLIS, for her assistance with the literature search.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2022.12.037.

References

  1. Pelletier J, Gottlieb M, Long B, Perkins Jr JC. necrotizing soft tissue infections (NSTI): pearls and pitfalls for the emergency clinician. J Emerg Med. 2022;62(4):480-91.
  2. Hersh AL, Chambers HF, Maselli JH, Gonzales R. National trends in ambulatory visits and Antibiotic prescribing for skin and soft tissue infections. Arch Intern Med. 2008; 168:1585-91.
  3. McHenry CR, Piotrowski JJ, Petrinic D, Malangoni MA. Determinants of mortality for necrotizing soft-tissue infections. Ann Surg. 1995;221(5):558-65.
  4. Elliott DC, Kufera JA, Myers RA. Necrotizing soft tissue infections. Risk factors for mortality and strategies for management. Ann Surg. 1996;224(5):672-83.
  5. Childers BJ, Potyondy LD, Nachreiner R, et al. Necrotizing fasciitis: a fourteen-year retrospective study of 163 consecutive patients. Am Surg. 2002;68(2):109-16.
  6. Stevens DL, Bryant AE. Necrotizing soft-tissue infections. N Engl J Med. 2017;377 (23):2253-65.
  7. Abdullah M, McWilliams B, Khan SU. Reliability of the laboratory risk Indicator in necrotising fasciitis (LRINEC) score. Surgeon. 2019;17:309-18.
  8. Keeling PA, Domes CM. A systematic review of the laboratory risk indicator for necrotizing fasciitis score and its utility in orthopedics for diagnosing necrotizing fasciitis in adults. Orthopedics. 2019;42:e288-94.
  9. Fernando SM, Tran A, Cheng W, Rochwerg B, Kyeremanteng K, Seely AJE, et al. Nec- rotizing soft tissue infection: diagnostic accuracy of physical examination, imaging, and LRINEC score: a systematic review and Meta-analysis. Ann Surg. 2019;269(1): 58-65.
  10. Gottlieb M, Avila J, Chottiner M, Peksa GD. Point-of-care ultrasonography for the di- agnosis of skin and soft tissue abscesses: a systematic review and meta-analysis. Ann Emerg Med. 2020;76(1):67-77.
  11. McInnes MDF, Moher D, Thombs BD, McGrath TA, Bossuyt PM, Clifford T, et al. Pre- ferred reporting items for a systematic review and meta-analysis of diagnostic test accuracy studies: the PRISMA-DTA statement. JAMA. 2018;319(4):388-96.
  12. Whiting PF, Rutjes AW, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al. QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy studies. Ann Intern Med. 2011;155(8):529-36.
  13. Bramer WM, Rethlefsen ML, Kleijnen J, Franco OH. Optimal database combinations for literature searches in systematic reviews: a prospective exploratory study. Syst Rev. 2017;6(1):245.
  14. Yen ZS, Wang HP, Ma HM, et al. Ultrasonographic screening of clinically-suspected necrotizing fasciitis. Acad Emerg Med. 2002;9:1448-51.
  15. Lin CN, Hsiao CT, Chang CP, et al. The relationship between fluid accumulation in ul- trasonography and the diagnosis and prognosis of patients with necrotizing fasciitis. Ultrasound Med Biol. 2019;45(7):1545-50.
  16. Lahham S, Shniter I, Desai M, et al. Point of Care Ultrasound in the diagnosis of necrotizing fasciitis. Am J Emerg Med. 2022;51:397-400.
  17. Castleberg E, Jenson N, Dinh VA. Diagnosis of necrotizing faciitis with bedside ultrasound: the STAFF Exam. West J Emerg Med. 2014;15(1):111-3.
  18. Parenti GC, Marri C, Calandra G, et al. Necrotizing fasciitis of Soft tissues: role of diagnostic imaging and review of the literature. Radiol Med. 2000;99:334-9.
  19. Chaudhry AA, Baker KS, Gould ES, Gupta R. Necrotizing fasciitis and its mimics: what radiologists need to know. AJR Am J Roentgenol. 2015;204(1):128-39.

Leave a Reply

Your email address will not be published. Required fields are marked *