International normalised ratio as an independent predictor of mortality in limb necrotising fasciitis with sepsis

Abstract

Introduction

Necrotising fasciitis with sepsis is a life threatening disease. The aim of this study was to analyse the association between international normalised ratio (INR) and mortality in sepsis patients with necrotising fasciitis.

Methods

A retrospective review was undertaken of 106 patients suffering from necrotising fasciitis with sepsis between November 2007 and December 2016. Data on comorbidities, clinical manifestations, laboratory findings, causative microbiological organisms, APACHE II (Acute Physiology and Chronic Health Evaluation II) score and outcomes were extracted. Logistic regression was carried out to examine the factors affecting mortality.

Results

Forty patients (37.7%) died. There was no significant difference in the white blood count (WBC) for the survivor and non-survivor groups. Non-survivors had a lower mean oxygenation index (OI) (288.7mmHg vs 329.4mmHg, p=0.032) and platelet count (PC) (139.5 vs 214.8 x 10⁹/l, p=0.028), and a higher mean INR (1.9 vs 1.3, p=0.000), activated partial thromboplastin time (APTT) (54.6 vs 44.2 seconds, p=0.005) and serum creatinine (2.3mg/dl vs 1.4mg/dl, p=0.007). Mortality in patients with INR >1.5 was significantly higher than in those with INR <1.5 when all risk factors (WBC, PC, OI, INR, APTT, creatinine) were considered (odds ratio: 4.414, 95% confidence interval: 1.263–15.428, p=0.020). Even after adjusting for age, sex, bacteraemia, diabetes and hepatic disorders, the data still exhibited elevated mortality for patients with INR >1.5 (odds ratio: 5.600, 95% confidence interval: 1.415–22.166, p=0.014).

Conclusions

INR is a significant independent predictor of mortality in sepsis patients diagnosed with necrotising fasciitis.

Introduction

Necrotising fasciitis (NF) is a severe soft tissue infection that causes extensive necrosis of subcutaneous tissue and fascia but mostly sparing the muscle and skin tissue. Despite recent improvement in the clinical care of patients with NF, mortality remains high. Multiple factors such as age, female sex, diabetes mellitus, bacteraemia and liver disease have been documented as contributing to NF related mortality.^1–3 Interval from admission to surgical debridement is the only potentially modifiable risk factor that has been identified in numerous studies. Early recognition of patients at increased risk of NF allows earlier and more aggressive care. In particular, early surgical intervention is crucial in preventing mortality.⁴ Multiple laboratory and radiographic tests have been employed in the evaluation of NF but none are optimal and all may lead to delays in therapy. Currently, the gold standard for diagnosis is operative exploration of the suspected area.

International normalised ratio (INR) is a laboratory parameter that measures the clotting of blood. It is used to determine the effects of oral anticoagulants on the clotting system. INR is therefore routinely used to evaluate the coagulation status of a patient for adjusting therapy. Furthermore, elevated INR is associated with severe morbidity and mortality in the presence of liver disease. However, few previous studies have examined the association between NF and INR. The aim of our retrospective study was to investigate the association between INR and NF at Wenzhou Medical University in China using data spanning a period of nine years.

Methods

Patients diagnosed with NF and sepsis in the upper or lower limbs between November 2007 and December 2016 were included in the study. Diagnosis of NF was based on intraoperative findings (Fig 1). Sepsis was defined using the criteria from the international sepsis conference in 2016, also known as the ‘sepsis-3’ definition.⁵ Patients were excluded from the study if they were aged <18 years, were taking any form of anticoagulation before admission, had any underlying haematological disorders or refused treatment. In addition, patients given low molecular weight heparin, warfarin or unfractionated heparin during their admission before taking blood samples were excluded from the analysis.

Figure 1.

Lower limbs of a patient infected by Staphylococcus leading to fascial necrosis and sepsis

Age, sex, presentation, duration of symptoms, underlying chronic diseases, causative microbiological organisms and laboratory parameters (white blood count [WBC], platelet count [PC], oxygenation index [OI], INR, activated partial thromboplastin time [APTT], serum creatinine) were reviewed. The severity of illness on admission was assessed using the APACHE II (Acute Physiology and Chronic Health Evaluation II) scoring system.

Statistical analysis

The data were analysed using SPSS^® version 15.0 (SPSS, Chicago, IL, US). Baseline demographic characteristics, laboratory findings, clinical presentation and bacteriology were compared between survivors and non-survivors. The significance of differences between groups in continuous variables was tested using either Student’s t-test or the Mann–Whitney U test. Categorical variables were analysed with either the chi-squared test or Fisher’s exact test. Multivariate logistic regression was used to evaluate the association between the diagnostic criteria of sepsis and death.

Results

The baseline characteristics of the 106 patients included in the study are summarised in Table 1. The mean patient age was 58.3 years (standard deviation [SD]: 13.8 years, range: 19–83 years) and 70% of the patients were male. Underlying chronic diseases included diabetes mellitus (n=48, 45%), liver cirrhosis (n=49, 46%), malignancy (n=6, 6%) and immunocompromised status (n=5, 5%). The infected sites were the lower limbs (n=95, 90%) or the upper limbs (n=11, 10%). On admission, the mean APACHE II score for the whole study cohort was 18.7 (SD: 7.4).

Table 1.

Baseline characteristics of necrotising fasciitis patients with sepsis

	Survivors (n=66)	Non-survivors (n=40)	p-value
Mean age (years)	58.4 (SD: 13.4)	58.0 (SD: 14.7)	0.883
Male sex	45 (68%)	29 (73%)	0.639
Infection in lower limbs	56 (85%)	39 (98%)	0.082
Median duration of symptoms before admission (days)	2 (IQR: 1–7)	2 (IQR: 2–4)	0.876
Mean APACHE II score on admission	15.3 (SD: 5.3)	24.1 (SD: 7.0)	0.000
Coexisting conditions
Diabetes mellitus	29 (44%)	19 (48%)	0.721
Hepatic disorders	25 (38%)	24 (60%)	0.027
Immunocompromised status	2 (3%)	3 (8%)	0.562
Malignancy	5 (8%)	1 (3%)	0.508
Signs and symptoms on admission
Body temperature >38.3°C or <36°C	38 (58%)	25 (63%)	0.617
Blood pressure <90/60mmHg	28 (42%)	22 (55%)	0.209
Laboratory findings on admission
Mean WBC (x 109/l)	13.2 (SD: 7.1)	11.7 (SD: 9.9)	0.402
Mean PC (x 109/l)	214.8 (SD: 176.1)	139.5 (SD: 154.8)	0.028
Mean OI (mmHg)	329.4 (SD: 95.8)	288.7 (SD: 89.4)	0.032
Mean INR	1.3 (SD: 0.3)	1.9 (SD: 0.9)	0.000
Mean APTT (seconds)	44.2 (SD: 11.4)	54.6 (SD: 20.5)	0.005
Mean creatinine (mg/dl)	1.4 (SD: 1.7)	2.3 (SD: 1.6)	0.007
Outcomes
Amputation	4 (6%)	2 (5%)	1.000
Mean length of hospital stay (days)	41.4 (SD: 19.3)	11.4 (SD: 13.9)	0.000

APACHE = Acute Physiology and Chronic Health Evaluation; APTT = activated partial thromboplastin time; INR = international normalised ratio; IQR = interquartile range; OI = oxygenation index; PC = platelet count; SD = standard deviation; WBC = white blood count

The prevalence of hepatic disorders was higher in non-survivors than in survivors (60% vs 38%, p=0.027). WBC was not significantly different between surviving and non-surviving patients. Non-survivors presented with a lower mean PC (139.5 [SD: 154.8] vs 214.8 [SD: 176.1] x 10⁹/l, p=0.028) and OI (288.7mmHg [SD: 89.4mmHg] vs 329.4mmHg [SD: 95.8mmHg], p=0.032), and a higher mean INR (1.9 [SD: 0.9] vs 1.3 [SD: 0.3], p=0.000), APTT (54.6 [SD: 20.5] vs 44.2 [SD: 11.4] seconds, p=0.005) and creatinine (2.3mg/dl [SD: 1.6mg/dl] vs 1.4mg/dl [SD: 1.7mg/dl], p=0.007). Survivors had a lower mean APACHE II score (15.3 [SD: 5.3] vs 24.1 [SD: 7.0], p=0.000). The mean length of hospital stay was longer in the survivor group than in the non-survivor group (41.4 [SD: 19.3] vs 11.4 [SD: 13.9] days, p=0.000). Forty patients died, giving an overall mortality rate of 37.7%.

Microbiological findings

Monobacterial Gram negative organisms were the most frequently identified (30.2%), followed by monobacterial Gram positive organisms (10.4%) and fungus (6.6%) (Table 2). Among Gram negative bacteria, Vibrio (17.9%) and Proteus (13.2%) were the most common organisms. Staphylococcus (14.2%) was the organism most often identified among Gram positive bacteria. In non-survivors, the presence of positive blood cultures was significantly higher than in survivors (42.5% vs 19.7%, p=0.012). The results of the blood cultures were: Vibrio vulnificus (n=10, 9.4%), Staphylococcus (n=8, 7.5%), Klebsiella (n=4, 3.8%), Candida (n=3, 2.8%), Acinetobacter baumannii (n=2, 1.9%), Proteus (n=2, 1.9%), and Streptococcus pyogenes, Aeromonas hydrophila, Escherichia coli, Pseudomonas aeruginosa, Corynebacterium, Klebsiella aerogenes and Shewanella algae. (all n=1, 0.9%).

Table 2.

Micro-organisms involved in necrotising fasciitis

	Overall (n=106)	Survivors (n=66)	Non-survivors (n=40)	p-value
Positive wound culture	63 (59.4%)	40 (60.6%)	23 (57.5%)	0.752
Positive blood culture (bacteraemia)	30 (28.3%)	13 (19.7%)	17 (42.5%)	0.012
Polybacterial infection	28 (26.4%)	20 (30.3%)	8 (20.0%)	0.244
Monobacterial Gram positive	11 (10.4%)	6 (9.1%)	5 (12.5%)	0.819
Monobacterial Gram negative	32 (30.2%)	18 (27.3%)	14 (35.0%)	0.401
Fungus	7 (6.6%)	2 (3.0%)	5 (12.5%)	0.134
Gram positive
Staphylococcus	15 (14.2%)	12 (18.2%)	3 (7.5%)	0.126
Streptococcus	8 (7.5%)	6 (9.1%)	2 (5.0%)	0.694
Enterococcus	9 (8.5%)	6 (9.1%)	3 (7.5%)	1.000
Gram negative
Vibrio	19 (17.9%)	9 (13.6%)	10 (25.0%)	0.139
Proteus	14 (13.2%)	11 (16.7%)	3 (7.5%)	0.177
Enterobacter	10 (9.4%)	8 (12.1%)	2 (5.0%)	0.383
Klebsiella	10 (9.4%)	7 (10.6%)	3 (7.5%)	0.851
Pseudomonas	8 (7.5%)	5 (7.6%)	3 (7.5%)	1.000
Acinetobacter	7 (6.6%)	5 (7.6%)	2 (5.0%)	0.909

Multivariate analysis for the major predictors of mortality

The patients were divided into high and low risk groups for each parameter according to their sequential organ failure assessment (SOFA)⁶ score and diagnostic criteria for sepsis.⁷ High risk was defined as SOFA score ≥2 (PC ≤100 x 10⁹/l, OI ≤300mmHg, creatinine ≥2mg/dl), INR >1.5, WBC <4 x 10⁹/l or >12 x 10⁹/l and APTT >60 seconds. The mortality in the higher risk INR group was significantly higher than in the lower risk INR group (odds ratio: 4.414, 95% confidence interval: 1.263–15.428, p=0.020) when all risk factors (WBC, PC, OI, INR, APTT, creatinine) were considered. WBC, PC, OI, APTT and creatinine were not significantly associated with mortality. Even when adjusting for age, sex, age, bacteraemia, diabetes mellitus and hepatic disorders, the data still exhibited elevated mortality for the higher risk INR group (odds ratio: 5.600, 95% confidence interval: 1.415–22.166, p=0.014) (Table 3).

Table 3.

Multivariate analysis for the major predictors of mortality

	Model 1*		Model 2†
	OR (95% CI)	p-value	OR (95% CI)	p-value
WBC	0.785 (0.309–1.994)	0.611	1.032 (0.369–2.880)	0.953
PC	0.855 (0.252–2.906)	0.802	1.394 (0.312–6.233)	0.664
OI	1.419 (0.566–3.563)	0.456	1.205 (0.456–3.182)	0.707
INR	4.414 (1.263–15.428)	0.020	5.600 (1.415–22.166)	0.014
APTT	2.028 (0.574–7.168)	0.273	2.284 (0.550–9.486)	0.256
Creatinine	1.688 (0.629–4.535)	0.299	1.922 (0.687–5.380)	0.214

APTT = activated partial thromboplastin time; CI = confidence interval; INR = international normalised ratio; OI = oxygenation index; OR = odds ratio; PC = platelet count; WBC = white blood count

*Adjusted for WBC, PC, OI, INR, APTT and creatinine

^†Further adjusted for age, sex, bacteraemia, diabetes mellitus and hepatic disorders

Discussion

This study analysed the association between INR and mortality in sepsis patients with limb NF. Sepsis is defined as life threatening organ dysfunction caused by a dysregulated host response to infection.⁵ A literature search identified ten studies published between 2005 and 2016 looking at mortality in NF patients (Table 4).^8–17 WBC, PC, INR, APTT and creatinine appeared to be the main risk factors for NF although there is no consensus among these studies for specific predictors of mortality. WBC, PC and creatinine do not have uniform criteria for evaluation of disease and prognosis.

Table 4.

Summary of studies investigating mortality in patients with necrotising fasciitis

Paper	Country	Study period	Mortality	Site of infection	WBC	PC	OI	INR	APTT	Creatinine
Lee, 20168	Taiwan	2004–2012	66.7% (total n=18)	Limbs	–	<150 x 109/l	–	–	–	–
Khamnuan, 20159	Thailand	2009–2012	19.3% (total n=1,504)	Limbs, Fournier gangrene, head and neck, trunk, multiple sites	–	–	–	–	–	≥1.6mg/dl
Krieg, 201410	Germany	1996–2011	32.8% (total n=64)	Limbs, perineum, trunk	–	–	–	–	–	≥4mg/dl
Lee, 201411	Taiwan	1996–2011	18% (total n=100)	Limbs, multiple sites	<10 x 109/l	<100 x 109/l	–	–	–	≥1.3mg/dl
Lin, 201312	Taiwan	2005–2011	16% (total n=75)	Limbs, trunk	<10 x 109/l	<100 x 109/l	–	>1.5	–	–
Yu, 201313	South Korea	2001–2012	33.7% (total n=83)	Limbs, buttocks, head and neck, trunk	–	<50 x 109/l	–	–	–	–
Huang, 201114	Taiwan	2003–2009	12.1% (total n=472)	Limbs, perineum, head and neck, trunk	Band polymorphonuclear neutrophils >10%	–	–	–	≥60 seconds	≥2mg/dl
Yeung, 201115	Hong Kong	Not known	28% (total n=29)	Limbs	–	<150 x 109/l	–	–	–	≥1.5 x normal
Tsai, 201016	Taiwan	2002–2008	25.7% (total n=70)	Limbs	Increased band forms of leucocytes	<80 x 109/l	–	–	–	–
Anaya, 200517	US	1996–2001	16.9% (total n=166)	Limbs, perineum, buttocks, head and neck, trunk	>30 x 109/l	–	–	–	–	≥2mg/dl

APTT = activated partial thromboplastin time; INR = international normalised ratio; OI = oxygenation index; PC = platelet count; WBC = white blood count

OI was not analysed in these studies but hyperbaric oxygen therapy is an effective tool in the treatment of NF.¹⁸ On univariate analysis of our cohort of NF patients presenting to the emergency department, OI was found to significantly influence mortality. We therefore feel that OI should be considered as a potential major risk factor together with other risk factors that have been well documented previously (WBC, PC, INR, APTT and creatinine). The correlation between these risk factors was also examined, and INR was shown to be significantly and independently associated with risk of mortality. To our knowledge, this is the first study to examine the correlation between WBC, PC, OI, INR, APTT and creatinine.

Based on our review of the existing literature, there are several possible explanations for our findings. Owing to the post-thrombolytic coagulation seen in acute ischaemic stroke,¹⁹ the INR is elevated, which renders the patient prone to excessive bleeding. It has been estimated that the risk of bleeding almost doubles for each one-point increase in INR above 3.0,²⁰ with the risk of intracranial haemorrhage similarly doubling for every unit rise in INR.²¹

While antimicrobial therapy and supportive treatment are essential early interventions when dealing with NF, surgery is thought to be key to improving the survival rate. When the wound is infected, it can release mass toxins and enzymes. This leads to tissue damage and necrotic tissue is an ideal breeding ground for bacteria. Consequently, the infection becomes more serious, leading to a vicious cycle. Debridement should be performed immediately because the time taken from admission to surgical intervention is related to survival rate in patients with NF.⁴ The insufficiency of coagulation of the blood increases the possibility of severe bleeding during the operation and patients have a higher chance of losing their lives.

An elevated INR might also indicate the presence of fatigue, in which inflammation has been reported to play a role.²² NF is a life threatening soft tissue infection that is accompanied by a rapid increase in inflammation and necrosis of the skin, which can progress to sepsis. Sepsis is commonly associated with a net procoagulant state and secondary consumptive conditions of thrombocytopenia. Large amounts of cytokines are released into the circulation, including tumour necrosis factor, interleukin 1 and interleukin 6.²³ These cytokines activate the coagulation system and affect anticoagulant mechanisms. As a result, microvascular clots are deposited, which may lead to tissue ischaemia and organ dysfunction.

If too many coagulation factors and platelets are consumed, bleeding is difficult to stop. This phenomenon is known as disseminated intravascular coagulation, which manifests itself as widespread microvascular thrombosis and haemorrhage. An elevated INR may therefore be associated with a greater severity of sepsis, which may explain the increased risk of mortality.

The SOFA score is a useful tool to assess the severity of sepsis. It based on six variables, representing coagulation, liver and kidney function, and the respiratory, cardiovascular and neurological systems. In addition, the laboratory risk indicator for necrotising fasciitis (LRINEC) is a robust scoring system capable of detecting NF even at an early stage.²⁴ Unfortunately, neither the SOFA score nor the LRINEC score incorporate INR.

Study limitations

One of the major limitations of our study is the retrospective study design. Medical records are not designed for the purposes of research and do not include all variables of interest to researchers. Moreover, this study did not investigate and cannot prove the pathogenic mechanism for our observed association between elevated INR and mortality in patients with NF. Finally, our patients were from a single medical centre in Wenzhou, China. The value of INR in predicting clinical outcomes may be different in other populations of NF patients. Further prospective studies with larger patient cohorts from multiple centres are necessary to more accurately evaluate the prognostic role of elevated INR as a predictor of mortality in patients with NF.

Conclusions

The mortality rate for NF with sepsis is high. INR appears to be a significant and independent predictor of mortality in sepsis patients with NF. This is not only because of the association between INR and sepsis but also because of the role INR plays in abnormal coagulation, which increases the possibility of severe bleeding intraoperatively, resulting in higher mortality. INR appears to be a laboratory risk indicator of NF that aids in pointing towards the initiation of early surgery and predicting a higher risk of death. It is unfortunate that the SOFA and LRINEC scoring systems do not take account of INR.

Funding information

This article was supported by the Science and Technology Project of Wenzhou (Grant No. Y20160328), the science and technology innovation team project of Foshan (2015IT100062), the study on biological three-dimensional printing tissue repair technology and clinical application—the aspect of full-thickness vascularized skin biological three-dimensional printing tissue repair technology and clinical application (2017C01054).