Evaluation of Factors Associated with Death in Patients with Necrotizing Fasciitis of the Abdominal Wall

Abstract

Background:

Necrotizing fasciitis (NF) is a rapidly progressing, life-threatening soft tissue infection. The prognostic factors associated with death from abdominal wall NF are not well understood.

Methods:

The medical records of 61 patients with abdominal wall NF were reviewed retrospectively. Demographic, co-morbidity, laboratory, and clinical data were collected and compared for patients who survived and who did not.

Results:

Sixty-one patients met the inclusion criteria, with eight deaths (13.11%). Elevated blood urea nitrogen (BUN) concentration, elevated total bilirubin, and prolonged intensive care unit (ICU) stay were correlated with a greater risk of death.

Conclusions:

Presentation with elevated markers of liver and renal dysfunction, as well as prolonged post-operative ICU stay were associated with a higher risk of death in patients with NF of the abdominal wall.

Necrotizing fasciitis (NF) is a rapidly progressing condition that can have devastating consequences. The reported mortality rates associated with the diagnosis have ranged from 6% to as high as 76% [1, 2]. Although the disease is relatively rare, with an estimated incidence of only 1,000 cases annually in the U.S., recently, the rate has increased [3,4]. Necrotizing fasciitis can affect any area of the body; however, the extremities and perineal area are more commonly affected than are the chest or torso [5].

Chest and abdominal wall NF present a unique case for surgeons, as debridement options can be more limited than those for extremity infections. Management of abdominal wall NF requires emergency surgical debridement, often with multiple subsequent debridement procedures, broad-spectrum antibiotics, and intensive medical management by a multi-disciplinary team of clinicians [6]. Even with early recognition and appropriate management, poor outcomes are possible.

Several studies have attempted to identify predictors of death in patients with NF. A review of almost 1,400 patients found that age >60 y, declining functional status, septic shock, thrombocytopenia, and renal failure requiring emergency hemodialysis were associated with a higher mortality rate [7]. Cultured growth of anaerobic bacteria necessitating repeated surgical debridements [8], multiple co-morbid conditions and delayed presentation [9], and presentation to outlying facilities that required inter-hospital transfer [10] have been suggested to be related to death. Risk stratification tools, including the Laboratory Risk Indicator for Necrotizing Fasciitis (LRINEC) and the Wall et al. criteria [11] have been developed to aid in the recognition of NF but offer little insight into prognosis.

Although several predictive variables have been postulated and identified for NF localized to the extremities, research into the prognosis of patients with abdominal wall NF is limited. Clinicians often are forced to make decisions in the setting of rapidly changing clinical pictures with unknown prognostic factors. A better understanding of mortality predictors, specifically for patients with NF involving the abdominal wall, will help guide treatment decisions in the future. The objective of this study was to identify prognostic factors associated with an increased risk of death in patients with NF of the abdominal wall.

Patients and Methods

Study design

This study evaluated factors associated with death in patients with NF of the abdominal wall. A 15-year retrospective chart view of such patients who were treated at two teaching hospitals in Southwest Michigan was completed. Institutional Review Board approval was obtained from each site included in the study.

Patient selection and data collection

All patients with the diagnosis of NF of the abdominal wall treated from January 1, 2000, to December 31, 2014, were identified using ICD-9 code (728.86 for NF) and CPT codes (11005 and 11006 for abdominal wall debridement). Inclusion criteria were age >18 y, admission because of abdominal wall NF, and treatment with surgical debridement.

Once patients were identified, medical records were reviewed retrospectively; demographic and clinical data were abstracted. The subjects were divided into two groups: Those who survived the infection (Group A) and those who did not (Group B). The groups were compared to assess differences in demographic factors, co-morbid conditions, days of care, number of debridement procedures, body mass index (BMI), admission vital signs, LRINEC score, and laboratory findings.

Statistical analysis

Independent statistical analysis was performed using SAS v. 9.4 for Fischer exact tests for categorical data and Mann Whitney tests for quantitative data, with significance set at p ≤ 0.05.

Results

Sixty-one patients with a diagnosis of NF of the abdominal wall were identified as fitting the inclusion criteria. Group A consisted of 53 patients, and Group B consisted of 8 patients (mortality rate 13.1%). Of the 61 patients, 34 were female (55.7%). The two patient groups were largely similar in the co-morbid conditions examined. Table 1 summarizes the demographic information and co-morbid conditions represented in the cohort.

Table 1.

Demographics and Co-Morbidities of 61 Patients with Necrotizing Fasciitis of the Abdominal Wall Treated between 2000 and 2014

	Total (%)	Group A (%)	Group B (%)
	61	53 (86.9)	8 (13.1)
Characteristic
Sex (%female)	34 (55.74)	32 (60.4)	2 (25.0)
Tobacco use (current)	27 (45)^a	30 (57.7)	3 (37.5)
Co-morbid conditions
Diabetes	33 (54.10)	27 (51.0)	6 (75)
Chronic obstructive pulmonary disease	8 (13.11)	7 (13.2)	1 (12.5)
Congestive heart failure	9 (14.75)	7 (13.2)	2 (25.0)
Peripheral vascular disease	4 (6.56)	4 (7.5)	0
Liver disease (hepatitis and cirrhosis)	1 (1.64)	0	1 (12.5)
Cancer	8 (13.11)	6 (11.3)	2 (25.0)
Two or more of the above conditions	27 (44.26)	22 (41.5)	5 (62.5)
Steroid use (longer than six mos)	4 (6.56)	2 (3.8)	2 (25.0)

N = 60.

Following analysis of laboratory markers and prognostic factors, three factors associated with a greater risk of death were identified. Group A had an average total bilirubin concentration of 0.85 mg/dL, which was significantly lower (p = 0.003) than in Group B, which had an average bilirubin concentration of 1.96 mg/dL. Group A experienced an average blood urea nitrogen (BUN) concentration of 23.92 mg/dL which was significantly lower (p = 0.002) than Group B, which averaged 57.50 mg/dL. Patients in Group A spent an average of 4.78 d under care in the intensive care unit (ICU), which was significantly less (p = 0.012) than patients in Group B, who spent 11.60 d on average in the ICU.

Laboratory values that were similar in the two groups were albumin concentration, white blood cell count, glucose concentration, hemoglobin, hematocrit, and total protein concentration. Although the values were not significantly different, patients in Group A were on average younger than patients in Group B (51.0 vs, 59.63 y) (p = 0.053). The average BMI of Group A was 40.36 kg/m² compared with 33.80 kg/m² in Group B (p = 0.063). Additional factors that were analyzed can be found in Table 2.

Table 2.

Laboratory and Clinical Data in Patients with Necrotizing Fasciitis of the Abdominal Wall Treated between 2000 and 2014 (Mean Values [Standard Deviation])

	Group A	Group B	p	Number of patients (n)
Age (years)	51.04 (12.97)	59.63 (8.73)	0.053	57
Body mass index (kg/m²)	40.36 (10.90)	33.80 (10.45)	0.063	63
Albumin (g/dL)	2.52 (0.60)	2.30 (0.63)	0.416	63
White blood cell count (cells × 10⁹/L)	17.77 (8.12)	15.00 (5.14)	0.358	63
Glucose (mg/dL)	170.83 (109.96)	174.50 (84.97)	0.724	62
Hemoglobin (g/dL)	11.77 (1.92)	11.35 (2.76)	0.749	55
Hematocrit (%)	34.79 (5.63)	35.59 (6.23)	0.738	62
Total protein (g/dL)	5.92 (1.14)	5.70 (1.26)	0.752	63
Calcium (mg/dL)	8.29 (0.84)	7.99 (0.78)	0.130	63
Sodium (mEq/L)	134.13 (4.67)	134.63 (5.85)	0.797	63
Chloride (mEq/L)	119.68 (152.20)	101.38 (8.63)	0.453	63
Potassium (mEq/L)	3.96 (0.54)	4.29 (0.88)	0.480	63
Bicarbonate (mEq/L)	22.59 (4.90)	20.00 (4.17)	0.122	59
Bilirubin (mg/dL)	0.85 (0.77)	1.96 (1.25)	0.003	58
AST (U/L)	35.93 (26.05)	41.14 (27.24)	0.554	55
ALT (U/L)	31.24 (22.18)	20.29 (9.74)	0.301	55
Alk phosphate (U/L)	99.51 (39.21)	94.00 (32.86)	0.757	55
Creatinine (mg/dL)	1.38 (0.87)	1.74 (1.32)	0.708	63
BUN (mg/dL) =	23.92 (14.97)	57.50 (29.99)	0.002	63
Temp average (°F)	98.82 (1.44)	98.30 (1.62)	0.428	59
Temp maximum (°F)	99.91 (1.83)	98.84 (1.83)	0.198	59
APACHE II score	16.17 (5.34)	20.00 (4.90)	0.178	24
LRINEC	8.00 (7.00)	9.00 (0.00)	1.00	5
Days of inpatient care	20.85 (16.86)	22.75 (28.54)	0.494	63
Days of intensive care unit care	4.78 (6.83)	11.60 (7.37)	0.012	63
Hours to debridement	61.60 (127.83)	14.67 (16.73)	0.063	58
Number of debridements	2.94 (1.92)	5.50 (8.77)	0.630	63

Group A consisted of patients who survived the infection and Group B of patients who did not.

Boldface values denote differences that are statistically significant (p < 0.05).

For abbreviations, see text.

Serum electrolytes examined were calcium, sodium, chloride, potassium, and bicarbonate. Overall, none of these values was significantly different in Group A and Group B. Liver function was examined using aspartate aminotransferase (AST), alanine aminotransferase (ALT), and alkaline phosphate (ALP). Groups A and B showed no statistically significant differences (p = 0.554, 0.301, and 0.757, respectively). Renal function was also examined, with Group A experiencing an average serum creatinine concentration of 1.38 mg/dL compared with Group B at 1.74 mg/dL (p = 0.708).

Body temperature, an indicator of infection, was similar in the two groups. Group A had an average total temperature of 98.82°F and an average maximum temperature of 99.91°F. When compared with Group B, which had an average total temperature of 98.30°F and an average maximum temperature of 98.84°F, no statistically significant difference was observed (p = 0.428 and 0.198, respectively). Acute Physiology and Chronic Health Evaluation (APACHE II) scores are measures of disease severity utilized in the ICU. The APACHE II scores for Groups A and B were 16.17 and 20.00, respectively (p = 0.178). The LRINEC scores were calculated utilizing laboratory values to distinguish between NF and severe cellulitis and were 8.00 and 9.00 for Groups A and B, respectively (p = 1.00).

Additional factors that may impact death in NF patients include length of time spent in in-patient care, hours until first surgical debridement, and the number of debridements. Group A spent an average of 20.85 d in total inpatient care, and Group B spent an average of 22.75 d (p = 0.494). Patients in Group A had an average of 61.60 h h until their first surgical debridement compared with 14.67 h in Group B (p = 0.063). Patients in Group A had an average of 2.94 debridements compared with 5.50 in Group B (p = 0.630).

Additional analysis of our results indicated that there was no influence in death or outcome regarding the site of the infection, causative agent of the infection, if it was single organism or polymicrobial, the antibiotic choice, or the type of wound care provided. The results can be seen in Table 3.

Table 3.

Comparison of Pathogen, Wound Location, Treatment, and Number of Antibiotics Given

	Group A	Group B	p	No. of patients
Median no. of debridements (range)	2.00 (7.00)	3.00 (27.00)	0.62	61
Median h to debridement (range)	24.00 (717.00)	8.50 (44.00)	0.06	56
Median no. of micro-organisms (range)	2.00 (4.00)	1.50 (4.00)	0.96	61
Median no. of antibiotics (range)	3.00 (7.00)	3.00 (5.00)	0.66	61
No. having >3 antibiotics – (%)	21 (39.62)	4 (50.00)	0.83	61
No. having >5 antibiotics (%)	3 (5.66)	1 (12.50)	0.92	61
Organism (%)
Enterococcus	6 (11.32)	2 (25.00)	0.94	61
Vancomycin-resistant Enterococcus	1 (1.89)	0 (0)	0.87	61
Bacteroides	12 (22.64)	1 (12.50)	0.45	61
Klebsiella	4 (7.55)	0 (0)	0.56	61
Proteus	3 (5.66)	1 (12.50)	0.92	61
Streptococcus	12 (22.64)	2 (25.00)	0.74	61
Escherichia coli	8 (15.09)	2 (25.00)	0.88	61
Lactobacilli	1 (1.89)	0 (0)	0.87	61
Methicillin-resistant Staphylococcus aureus	4 (7.55)	0 (0)	0.56	61
Other Staphylococcus	6 (11.32)	1 (12.50)	0.77	61
Unidentified anaerobes	10 (18.87)	0 (0)	0.22	61
Propionibacterium	4 (7.55)	0 (0)	0.56	61
Candida	2 (3.77)	0 (0)	0.75	61
Prevotella	5 (9.43)	0 (0)	0.48	61
Corynebacterium	4 (7.55)	1 (12.50)	0.88	61
Pseudomonas	1 (1.89)	2 (25.00)	1.00	61
Clostridium perfringens	1 (50.00)	1 (100.00)	1.00	3
Wound treatment (%)
Negative-pressure therapy	34 (64.15)	3 (37.50)	0.15	61
Whirlpool	2 (3.77)	0 (0)	0.75	61
Wet to dry	6 (11.32)	1 (12.50)	0.77	61
Other	14 (26.42)	5 (62.50)	0.99	61
Wound location (%)
Flank	1 (2.13)	1 (12.50)	0.98	55
Perineum	17 (36.17)	3 (37.50)	0.69	55
Abdomen	19 (40.43)	2 (25.00)	0.34	55
Breast/axilla	4 (8.51)	0 (0)	0.52	55
Buttocks	2 (4.26)	0 (0)	0.73	55
Back	2 (4.26)	0 (0)	0.73	55
Inguinal/leg	2 (4.26)	1 (12.50)	0.95	55

Discussion

In this study, prognostic factors associated with death in patients with NF of the abdominal wall were evaluated. Necrotizing fasciitis is a relatively rare disease, and abdominal wall NF represents only a portion of the total cases. Abdominal wall NF rates have been reported to be as low 8.5% [12] and as high as 18% [13] of NF involving all body sites. Successful recognition, treatment, and management of these cases is difficult but vital to maximize survival.

Whereas surgical debridement and antibiotic therapy are the mainstays of treatment, markers of prognosis are less well defined. In this study, elevated bilirubin, elevated BUN, and prolonged stay in the ICU emerged as important markers of a poor prognosis. Liu et al. also noted that renal injury marked by elevated BUN and creatinine concentrations correlated with a higher risk of death in patients with NF in various body sites [14]. Following debridement and medical management with antimicrobial agents, prolonged stay in the ICU is correlated with a higher risk of death. Gunter et al. also found that longer ICU stay was associated with an increased risk of death in a study of necrotizing soft tissue infections [15]. In this study, similar markers of prognosis were identified for abdominal wall NF as had been identified for NF affecting more peripheral sites.

Whereas several factors, including age, hours to debridement, and the number of debridements, were not significantly different in our two groups, there does appear to be a trend in that the patients in Group B who succumbed to infection were older, required debridement more quickly, and received more debridements. It is possible that with greater patient numbers, these factors will become significant. Thus, the data suggest that, not surprisingly, the patients who succumbed to infection had more extensive NF than patients who survived the infection. However, proof of this idea would require measurements of the infected sites. Our data support the idea that serious infections require aggressive surgical debridement, wound management, and targeted antibiotics to overcome them.

Future studies on NF of the abdominal wall can further our understanding of the prognosis of these patients. Our study, despite researching records over a period of 15 years, had a relatively small sample. Our results were limited by older records lacking completeness and our inability to calculate LRINEC scores because of the low utilization of C-reactive protein measurements in the early 2000s. Further research into the exact pathology underlying the elevated BUN concentration also could be beneficial to understanding the whole clinical picture.

Conclusion

In this study, subjects with NF of the abdominal wall who died had significantly higher bilirubin and BUN concentrations and significantly longer ICU stays than subjects who survived. With a better understanding of the prognosis of patients with NF, appropriate aggressiveness of treatment plans can be adjusted more aptly to the needs of the particular patient. In addition, patients, their families, and their medical providers will be better equipped to make decisions regarding care. Laboratory data, specifically bilirubin and BUN measurements, can help determine the advancement of the disease; and prolonged stay in the ICU is an ominous sign in the post-operative period.

Footnotes

Acknowledgments

The authors thank Western Michigan University Homer Stryker M.D. School of Medicine Department of Epidemiology and Biostatistics for their assistance with this project.

Author Disclosure Statement

The authors have no financial relationships to disclose.

References

Sudarsky

, Laschinger

, Coppa

, Spencer

. Improved results from a standardized approach in treating patients with necrotizing fasciitis. Ann Surg, 1987; 206:661–665.

Stone

, Martin

Jr . Synergistic necrotizing cellulitis. Ann Surg, 1972; 175:702–711.

Ellis Simonsen

, Van Orman

, Hatch

, et al. Cellulitis incidence in a defined population. Epidemiol Infect, 2006; 134:293–299.

Salcido

. Necrotizing fasciitis: Reviewing the causes and treatment strategies. Adv Skin Wound Care, 2007; 20:288–293.

Hakkarainen

, Kopari

, Pham

, Evans

. Necrotizing soft tissue infections: Review and current concepts in treatment, systems of care, and outcomes. Curr Probl Surg, 2014; 51:344–362.

Roje

, Roje

, Matic

, et al. Expression of concern: Necrotizing fasciitis, literature review of contemporary strategies for diagnosing and management with three case reports: Torso, abdominal wall, upper and lower limbs. World J Emerg Surg, 2011; 6:46.

Faraklas

, Stoddard

, Neumayer

, Cochran

. Development and validation of a necrotizing soft-tissue infection mortality risk calculator using NSQIP. J Am Coll Surg, 2013; 217:153–160.

Krieg

, Röhrborn

, Schulte Am Esch

, et al. Necrotizing fasciitis: Microbiological characteristics and predictors of postoperative outcome. Eur J Med Res, 2009; 14:30–36.

Wong

, Chang

, Pasupathy

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

10.

Holena

, Mills

, Carr

, et al. Transfer status: A risk factor for mortality in patients with necrotizing fasciitis. Surgery, 2011; 150:363–370.

11.

Wall

, de Virgilio

, Black

, Klein

. Objective criteria may assist in distinguishing necrotizing fasciitis from nonnecrotizing soft tissue infection. Am J Surg, 2000; 179:17–20.

12.

Shaikh

, El-Menyar

, Natesa Mudali

, et al. Clinical presentations and outcomes of necrotizing fasciitis in males and females over a 13 year period. Ann Med Surg, 2015; 4:355–360.

13.

Golger

, Ching

, Goldsmith

, et al. Mortality in patients with necrotizing fasciitis. Plast Reconstr Surg, 2007; 119:1803–1807.

14.

Liu

, Chi

, Ho

, et al. Microbiology and factors affecting mortality in necrotizing fasciitis. J Microbiol Immunol Infect, 2005; 38:430–435.

15.

Gunter

, Guillamondegui

, May

, Diaz

. Outcome of necrotizing skin and soft tissue infections. Surg Infect, 2008; 9:443–450.