Necrotizing Soft Tissue Infection: Time is Crucial,and the Admitting Service Matters

Abstract

Background:

Early diagnosis and prompt debridement of necrotizing soft tissue infection (NSTI) improves the outcome. We sought to determine whether failure to admit NSTI patients to acute care surgery (ACS) departments delays treatment and increases the mortality rate.

Methods:

Patients with NSTI were identified using the 2007–2018 institutional emergency surgery database at a tertiary care hospital. The diagnosis was confirmed by the operative/pathology reports. Patients who developed NSTI during hospitalization or underwent initial debridement at an outside hospital were excluded. Patients admitted to a non-ACS service (e.g., medicine, gynecology) were compared with those admitted to the ACS service with respect to co-morbidities, clinical presentation, time to surgery, and mortality rate. Multi-variable linear and logistic analyses were performed to determine whether admission to a non-ACS service predicts a delay in surgery or an increase in the mortality rate.

Results:

Of 132 patients, 91 met the inclusion criteria. The mean age was 53 years; 56% were male. Twenty patients (22%) were admitted to a non-ACS service, two thirds of them with an initial misdiagnosis (e.g., cellulitis). The demographics, co-morbidities, and clinical presentation were similar in the two groups except that the non-ACS group more often had human immunodeficiency virus infection (15.0% versus 2.8%; p = 0.04) and less often presented with erythema (70% versus 94.4%; p = 0.01). The median time to incision in non-ACS patients was significantly longer (24.8 versus 3.9 hours; p < 0.001). The mortality rates were 20.0% for the non-ACS group and 7.0% for the ACS group (p = 0.086). Multi-variable analyses revealed that absence of erythema is independently associated with a non-ACS admission (odds ratio [OR] 5.9; 95% confidence interval [CI] 1.3–25.6; p = 0.02), and non-ACS admissions correlated independently with delayed surgery (OR 35.20; 95% CI 3.86–321.20; p = 0.002).

Conclusions:

Admission of patients with NSTI to a non-ACS service often occurs because of initial misdiagnosis, especially in the absence of skin erythema; correlates with significantly delayed surgery; and might lead to more deaths.

Necrotizing soft tissue infection (NSTI) often is characterized by rapidly progressive and extensive necrosis along soft tissue planes as well as systemic toxicity and multi-system organ dysfunction [1]. Depending on the location and the depth of disease involvement, NSTI was known historically by different names such as necrotizing fasciitis, gas gangrene, Ludwig angina, and Fournier gangrene. Despite all the scientific advancement in diagnostic techniques and quality of surgical care over the last few decades, the mortality rate remains elevated, ranging from 12% to 45% [2,3]. Several studies have identified risk factors for death in NSTI such as advanced age [4], co-morbidities (e.g., diabetes mellitus) [5], immunosuppression [6], intravenous drug use [7], and delay in initiation of surgical debridement or antibiotics [8]. Prompt and aggressive surgical debridement is the single most important factor that can rescue these patients; delay in surgical debridement can be detrimental [3,9 –11].

Keeping NSTI in the differential diagnosis of patients presenting with critical illness is of the utmost importance because of the disease's non-specific presentation in early stages [4,9,12], its rapid progression, and its high mortality rate if not diagnosed promptly and treated aggressively. Occasionally, the non-specific symptoms and physical examination (e.g., pain, erythema) result in misdiagnosis and admission to clinical services with little experience in recognizing and treating NSTI [13,14]. Surgeons trained in or practicing acute care surgery (ACS) are arguably the most experienced in recognizing and treating these patients. The impact on outcomes of admission of patients with NSTI to a non-surgical or a non-ACS service remains largely unknown. We investigated whether admission of NSTI patients to a non-ACS service results in delay in definitive treatment, a higher patient mortality rate, or both.

Patients and Methods

Patient population

The Emergency General Surgery Database, a locally created database maintained at the Massachusetts General Hospital, was reviewed to identify all patients with NSTI admitted from 2007–2018. The database was queried specifically for all admissions or consults with a post-operative diagnosis of NSTI, necrotizing fasciitis, gas gangrene, or Fournier gangrene using International Classification of Diseases (ICD)-9-CM codes 728.86, 040.0, and 608.83 and ICD-10-CM M72.6, A48.0, and N49.3. Patients were excluded if they were admitted for clearly unrelated diseases and developed NSTI during hospitalization, if their first debridement was performed at an outside hospital prior to transfer to our institution, or if the patients' goals of care included Do Not Resuscitate (DNR)/Do Not Intubate (DNI). Next, a systematic retrospective chart review was performed both to confirm the diagnosis by operative and pathologic reports and to collect additional data (e.g., admitting service, additional co-morbidities, clinical presentation and physical examination findings, and time to operation). The Laboratory Risk Indicator for Necrotizing fasciitis (LRINEC) score was calculated for patients having all necessary values available [15].

Admitting service

Two patient cohorts were created: Those admitted to the ACS service and those initially admitted to a non-ACS service (e.g., medicine, obstetrics, gynecology, and orthopaedics).

Primary outcome

The primary outcome of the study was the time to first surgical incision, which was defined as the time in hours from initial hospital presentation to the time of incision. Delayed surgery was defined as any first procedure not being performed within 4.6 hours, the median time to first incision in our study.

Secondary outcome

The secondary outcome of the study was in-hospital death.

Reasons for non-ACS admission

For each patient who was not admitted to the ACS service, we investigated the reason behind that decision, with a specific focus on whether there was an initial misdiagnosis.

Statistical analysis

Uni-variable then multi-variable analyses were performed to study whether admission to a non-ACS service resulted in a longer time to first surgical incision and a higher in-hospital mortality rate. Categorical variables are presented as totals and percentages. Continuous variables are presented as mean or median with interquartile range (IQR), depending on the data distribution. The χ² test was used for comparison of discrete variables. The Student t-test and Wilcoxon rank sum test were used, as appropriate, for the analysis of continuous variables. All analyses were performed with STATA 14.2 (Stata Corp, College Station, TX). The study was approved by Partners Human Research Institutional Review Board.

Results

During the study period, 132 patients with NSTI were identified in the database. Ninety-one patients met the inclusion criteria; 56% were male, the mean age was 53 years, and the mean body mass index was 29.7 kg/m². Figure 1 demonstrates that NSTI most often involved the lower extremities (52.8%). There was no difference in the NSTI location between the non-ACS and ACS patients.

FIG. 1.

Anatomic location of infection.

Table 1 describes the clinical signs and symptoms on initial presentation. In summary, most patients were afebrile, the mean systolic blood pressure was 116 mm Hg, and more than 80% of the patients had tachycardia (pulse >100/min.). Severe pain and erythema were the most common clinical symptom and sign, respectively, both at 90%. The classic pathognomonic signs of NSTI, such as hemorrhagic bullae and crepitus, were reported less often, at 14.4% and 12.2%, respectively. Patients often had hypoalbuminemia, with the median serum albumin concentration being 2.4 g/dL (Table 2). The median time to first incision of all patients was 4.6 hours, and the overall mortality rate was 9.9% (Table 3).

Table 1.

Demographic Characteristics and Clinical Presentation

Feature	Non-ACS admission (n = 20)	ACS admission (n = 71)	Total (n = 91)	p
Mean age (y)	51.4	53.7	53.2	0.54
Male (%)	15 (75)	36 (50.7)	51 (56.0)	0.05
Median BMI (IQR)	27.4 (24.1–35.1)	30.1 (24.3–36.6)	29.7 (24.3–36.3)	0.80
Mean systolic blood pressure at presentation mm Hg (± standard deviation)	116 (± 21)	116 (± 21)	116 (± 21)	0.98
Referral status (%)	15 (75)	48 (71.6)	61 (72.4)	0.77
Physical examination (%)
Fever (temperature >38°C)	8 (40)	17 (23.9)	25 (27.5)	0.15
Tachycardia (heart rate >100/min)	16 (80)	58 (81.7)	74 (81.3)	0.86
Septic shock^a	5 (25)	21 (29.6)	26 (28.6)	0.69
Pain	17 (89.5)	68 (95.8)	85 (94.4)	0.29
Erythema	14 (70.0)	67 (94.4)	81 (90.0)	0.01
Induration	7 (36.8)	29 (42.0)	36 (40.9)	0.68
Bullae	3 (15.8)	10 (14.1)	13 (14.4)	0.85
Crepitus	1 (5.3)	10 (14.1)	11 (12.2)	0.30
Co-morbidities (%)
Diabetes mellitus	15 (75)	48 (67.6)	63 (69.2)	0.53
Hypertension	6 (30)	30 (42.3)	36 (39.6)	0.32
HIV infection	3 (15)	2 (2.8)	5 (5.5)	0.04
Transplant status	1 (5)	2 (2.8)	3 (3.3)	0.63
Malignant disease^b	4 (2)	5 (7.0)	9 (4.9)	0.09
Intravenous drug abuse	4 (20)	9 (12.7)	13 (14.3)	0.41
Autoimmune disease	1 (5)	3 (4.2)	4 (4.4)	0.88
Steroid use	0	10 (14.1)	10 (11.0)	0.08
Peripheral vascular disease	0	1 (1.4)	1 (1.1)	0.59
Cirrhosis	0	2 (2.8)	2 (2.2)	0.45
Renal dysfunction requiring dialysis	0	3 (4.2)	3 (3.3)	0.35

Septic shock was defined as the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3)[ 22].

Malignant disease was defined as all solid tumor with metastasis or ongoing chemotherapy, including hematologic malignancy (lymphoma and leukemia).

BMI = Body Mass Index; HIV = human immunodeficiency virus.

Table 2.

Laboratory Values (Median [Interquartile Range]) Closest to Time of Surgery

Feature	Non-ACS admission (n = 20)	ACS admission (n = 71)	Total (n = 91)	p
WBC × 10³/mcL	16.8 (8.1–19.9)	14.7 (8.5–20.2)	15.4 (8.5–19.9)	0.86
Platelets × 10³/mcL	245 (153.5–289)	211 (149–311)	218 (150–294)	0.55
Serum BUN mg/dL	19.5 (11–31.5)	23 (12–40)	23 (11–36)	0.46
Serum creatinine mg/dL	1.2 (0.8–1.8)	1.3 (0.8, 2.0)	1.2 (0.8–2.0)	0.58
Serum sodium concentration mmol/L	134 (131.5–137)	135 (132–138)	135 (132–138)	0.38
Serum glucose concentration mg/dL	143 (107–222)	139 (105–193)	139 (106–193)	0.49
Serum C-reactive protein concentration mg/L	171.8 (146.9–276.5)	134.9 (115.8– 142.8)	138.2 (116.5–224.2)	0.04
Serum lactate concentration mmol/L	1.6 (1.3–3)	1.6 (1.1–2.1)	1.6 (1.1–2.1)	0.82
Serum albumin concentration g/dL	2.4 (2–3)	2.5 (2.1–2.9)	2.4 (2–2.9)	0.60
International normalized ratio	1.3 (1.2–1.5)	1.2 (1.1–1.4)	1.3 (1.2–1.4)	0.23
LRINEC score	6.5 (5–7)	5 (3.5–7.5)	5 (4–7)	0.26

BUN = blood urea nitrogen, LRINEC = Laboratory Risk Indicator for Necrotizing fasciitis; WBC = white blood cell count.

Table 3.

Treatment Variables and Outcomes

Variable	Non-ACS admission (n = 20)	ACS admission (n = 71)	Total (n = 91)	p
Median time from presentation to start of antibiotic (IQR) (h)	2.3 (1.3–7.1)	1.8 (1.0–3.4)	1.9 (1.1–3.6)	0.33
Median time from presentation to surgical incision (IQR) (h)	24.8 (9.0–33.7)	3.9 (2.6–7)	4.6 (2.9–9.0)	<0.001
No. with delay in surgical incision (> 4.6 h; median time from presentation to surgical incision) (%)	17 (94.4)	27 (40.9)	44 (52.4)	<0.001
No. of trips to operating room (%)	3 (1.5–5)	4 (2–7)	4 (2–7)	0.15
Post-operative length of stay, median (IQR) (d)	18 (8.5–29.5)	13 (9–26)	14 (9–26)	0.63
Median hospital length of stay (IQR) (d)	20 (9.5–30.5)	14 (9–26)	15 (9–28)	0.38
Amputations (%)	4 (20)	9 (12.7)	13 (14.3)	0.41
Deaths (%)	4 (20)	5 (7.0)	9 (9.9)	0.086

Admitting service

Twenty patients (22.0%) were admitted to non-ACS services. In this group, 13 patients (65%) initially had a misdiagnosis (e.g., cellulitis, septic arthritis, thrombophlebitis). Four patients (20%) had a correct diagnosis but were admitted to the medicine service because of their overwhelming concomitant co-morbidities, such as congestive heart failure or hematologic malignancy. Three patients (15%) suffered from NSTI as a surgical site complication and were admitted to the operating service, which then consulted ACS.

Uni-variable analyses

Table 1 shows the uni-variable analyses comparing the two cohorts of patients. In summary, the two groups were similar in demographics and co-morbidities, except that the non-ACS group more often had HIV infection (15% versus 2.8%; p = 0.04). The two cohorts also were similar in terms of clinical presentation, except that there was less reporting of skin erythema in the non-ACS group (70% versus 94.4%; p = 0.01). All laboratory findings indicating the severity of NSTI, including the calculated LRINEC score, were similar in the two cohorts (Table 2), except for the serum C-reactive protein (CRP) concentration, which was higher in the non-ACS group (172 mg/dL versus 135 mg/dL; p = 0.04).

Time to first surgical incisions

The time from presentation to first surgical incision was significantly greater for the non-ACS group (24.8 versus 3.9 hours; p < 0.001) (Fig. 2). The delay was predominantly prior to surgical consultation. For the patients with a misdiagnosis, the time to incision was even higher at 28.6 hours. The time to initiation of antibiotics after hospital presentation was similar at 1.9 hours (Table 3). The time to first dose of antibiotic was not different between the non-ACS and ACS patients (2.3 hours versus 1.8 hours; p = 0.33). The location of the NSTI did not correlate with any delay in surgery.

FIG. 2.

Time from hospital presentation to incision (h). Time was 24.8 h for non-acute care surgery (non-ACS) group compared with 3.9 h for ACS group. For patients with misdiagnosis, time to incision was even higher at 28.6 h. P-value comparing ACS with non-ACS admission.

Mortality rate

The mortality rate for the non-ACS group was 20% compared with 7% for the ACS group (Fig. 3), although this difference did not reach statistical significance (p = 0.086). For the patients with a misdiagnosis, the mortality rate was even higher at 23%. Other outcomes, such as the length of hospital stay, total number of operations, and amputation rates, were similar for the two groups.

FIG. 3.

Mortality rate for non-acute care surgery (non-ACS) group was 20% compared with 7% for ACS group, and rate was higher, 23%, for patients with misdiagnosis. P-value comparing ACS with non-ACS admission.

Multi-variable analyses

In multi-variable logistic analyses (Table 4), after adjusting for disease severity at presentation (septic shock, tachycardia, presence of crepitus) and intravenous drug use status, admission to a non-ACS service was associated with delayed surgery (odds ratio [OR] 35.20; 95% confidence interval [CI] 3.86–321.20; p = 0.002).

Table 4.

Multivariable Analysis of Factors Associated with Delayed Surgery

Variable	Adjusted odds ratio	95% Confidence interval	p
Non-ACS admission	35.2	3.86–321.20	0.002
Intravenous drug abuse	3.99	0.76–20.84	0.101
Tachycardia	0.09	0.15–0.50	0.006
Crepitus	0.15	0.20–1.08	0.059
Septic shock	0.44	0.11–1.71	0.238

In another multi-variable analysis where admission to a non-ACS service was treated as the dependent variable and adjusted for the HIV infection status, absence of erythema was independently associated with admission to a non-ACS service (OR 5.9; 95% CI 1.3–25.6; p = 0.02).

Discussion

To the best of our knowledge, our study is the first to suggest that: (1) Two thirds of the patients with NSTI admitted to a non-ACS service receive an initial misdiagnosis; (2) absence of skin erythema could be related to such misdiagnosis; (3) admission to a non-ACS service is correlated independently with a significant delay in definitive operative treatment; and (4) admission to a non-ACS service might increase the mortality rate, even though larger studies clearly are needed to demonstrate the latter finding statistically.

Necrotizing soft tissue infection is a rapidly progressive and life-threatening condition with mortality rates ranging from 12% to 45% [12]. The clinical presentation often is either atypical or non-specific, resembling more common and less severe diseases such as cellulitis and erysipelas [16], making its diagnosis challenging. Misdiagnosis has been described in 40%–90% of cases [12].

The misdiagnosis rate in our study was lower (13 patients; 14.3%) than in prior published reports [14,16], with all patients with misdiagnosis having been admitted to a non-ACS service. One possible reason is that our institution is a tertiary care center, with a large number of patients transferred from other hospitals or emergency rooms, sometimes with an erroneous diagnosis. Another possible explanation is the near-routine use of advanced diagnostic techniques such as computed tomography (CT) scans in our institution. Martinez et al. proposed that the sensitivity of CT scans in identifying NSTI is as high as 100% and the specificity is 98% [17], but many surgeons remain skeptical that the negative predictive value of CT scans really is that high, even with experienced reading radiologists.

Our main study finding is that failure to admit patients with NSTI to the ACS service has serious consequences. Specifically, in a disease in which every minute counts and surgery is the mainstay of rescue, the time to first incision was longer in patients admitted to a non-ACS service, from less than four hours to more than 24 hours. Such delay often comes because of failure to recognize NSTI promptly, resulting in late ACS consultation rather than a delay from ACS consultation to institution of surgery. The impact of this delay was a near tripling of the mortality rate from 7% to 20%, although that difference did not reach statistical significance, likely because of a type II error resulting from the small number of patients. Several prior studies have shown that delay in surgery in patients with NSTI is catastrophic, correlating directly with a higher mortality rate [11,16,18]. A similar impact on the outcome has been demonstrated in other emergency general surgery diagnoses, such as small bowel obstruction and acute mesenteric ischemia, where initial admission to a non-surgical service was associated with a delay in time to surgical intervention and higher rates of death [19 –21]; the impact of such admission in patients with NSTI, as demonstrated by our data, is even more deleterious. As such, quality improvement projects are needed to increase awareness among medical and emergency room providers of the frequent atypical presentation of NSTI and to encourage a lower threshold for consulting the ACS team early for patients presenting with atypical skin and soft tissue infections or associated systemic signs of the infection.

The overall mortality rate among all patients was 9.9%, which is lower than that in many prior studies. This could be explained in part by the earlier surgery in patients admitted to the ACS service. Specifically, the median time to surgery of only 4.6 hours is much lower than the mean of 12 to 45 hours described in most published studies [10,11].

Our study has a few major limitations. First, it is a retrospective single institution study, making generalizability difficult. Second, during chart review, we assumed that the absence of documentation was equivalent to the absence of the symptom or physical examination findings. Third, we did not examine the microbiology of NSTI nor the choice of antibiotics. Fourth, the study has a small sample, resulting in a likely type II error and in a less than optimal number of confounder adjustments in our multi-variable analyses.

Conclusion

Admission of NSTI patients to a non-ACS service often occurs because of initial misdiagnosis, especially in the absence of skin erythema on physical examination. The non-ACS service admission correlates with a significant time delay to definitive surgical debridement and might lead to a higher mortality rate. Admitting physicians should have a high-index of suspicion for NSTI and a low threshold for an early ACS service consultation for prompt evaluation and management.

Footnotes

Author Disclosure Statement

We have no conflicts of interest to report. We have no disclosures.

References

Kaafarani

, King

. Necrotizing skin and soft tissue infections. Surg Clin North Am, 2014; 94:155–163.

Mills

, Faraklas

, Davis

, et al. Outcomes from treatment of necrotizing soft-tissue infections: Results from the National Surgical Quality Improvement Program database. Am J Surg, 2010; 200:790–796.

Stevens

, Bryant

. Necrotizing soft-tissue infections. N Engl J Med, 2018; 378:971.

Wang

, Lim

. Necrotizing fasciitis: Eight-year experience and literature review. Braz J Infect Dis, 2014; 18:137–143.

Lee

, Kuo

, Peng

, et al. Prognostic factors and monomicrobial necrotizing fasciitis: Gram-positive versus gram-negative pathogens. BMC Infect Dis, 2011; 11:5.

Keung

, Liu

, Nuzhad

, et al. Immunocompromised status in patients with necrotizing soft-tissue infection. JAMA Surg, 2013; 148:419–426.

Chen

, Fullerton

, Flynn

. Necrotizing fasciitis associated with injection drug use. Clin Infect Dis, 2001; 33:6–15.

Tillou

, St Hill

, Brown

, Velmahos

. Necrotizing soft tissue infections: Improved outcomes with modern care. Am Surg, 2004; 70:841–844.

Sarani

, Strong

, Pascual

, Schwab

. Necrotizing fasciitis: Current concepts and review of the literature. J Am Coll Surg, 2009; 208:279–288.

10.

McHenry

, Piotrowski

, Petrinic

, Malangoni

. Determinants of mortality for necrotizing soft-tissue infections. Ann Surg, 1995; 221:558–565.

11.

Kobayashi

, Konstantinidis

, Shackelford

, et al. Necrotizing soft tissue infections: Delayed surgical treatment is associated with increased number of surgical debridements and morbidity. J Trauma, 2011; 71:1400–1405.

12.

Goh

, Goh

, Ang

, Wong

. Early diagnosis of necrotizing fasciitis. Br J Surg, 2014; 101:e119–e125.

13.

Lille

, Sato

, Engrav

, et al. Necrotizing soft tissue infections: Obstacles in diagnosis. J Am Coll Surg, 1996; 182:7–11.

14.

Frazee

, Fee

, Lynn

, et al. Community-acquired necrotizing soft tissue infections: A review of 122 cases presenting to a single emergency department over 12 years. J Emerg Med, 2008; 34:139–146.

15.

Wong

, Khin

, Heng

, et al. The LRINEC (Laboratory Risk Indicator for Necrotizing fasciitis) score: A tool for distinguishing necrotizing fasciitis from other soft tissue infections. Crit Care Med, 2004; 32:1535–1541.

16.

Elliott

, Kufera

, Myers

. Necrotizing soft tissue infections: Risk factors for mortality and strategies for management. Ann Surg, 1996; 224:672–683.

17.

Martinez

, Peponis

, Hage

, et al. The role of computed tomography in the diagnosis of necrotizing soft tissue infections. World J Surg, 2018; 42:82–87.

18.

Wong

, Chang

, Pasupathy

, et al. Necrotizing fasciitis: Clinical presentation, microbiology, and determinants of mortality. J Bone Joint Surg Am, 2003; 85:1454–1460.

19.

Aquina

, Becerra

, Probst

, et al. Patients with adhesive small bowel obstruction should be primarily managed by a surgical team. Ann Surg, 2016; 264:437–447.

20.

Oyasiji

, Angelo

, Kyriakides

, Helton

. Small bowel obstruction: Outcome and cost implications of admitting service. Am Surg, 2010; 76:687–691.

21.

Eltarawy

, Etman

, Zenati

, et al. Acute mesenteric ischemia: The importance of early surgical consultation. Am Surg, 2009; 75:212–219.