Biomarkers in post‐reperfusion syndrome after acute lower limb ischaemia

Abstract

Ischaemia reperfusion (I/R) injury refers to tissue damage caused when blood supply returns to the tissue after a period of ischaemia. Matrix metalloproteinases (MMPs), neutrophil gelatinase‐associated lipocalin (NGAL) and cytokines are biomarkers involved in several vascular complications. The aim of this study was to evaluate the role of MMPs, NGAL and inflammatory cytokines in I/R syndrome. We conducted an open label, multicentric, parallel group study, between January 2010 and December 2013. Patients with acute limb ischaemia were enrolled in this study and were divided into two groups: (i) those subjected to fasciotomy and (ii) those not subjected to fasciotomy, according to the onset of compartment syndrome. Plasma and tissue values of MMPs and NGAL as well as plasma cytokines were evaluated. MMPs, NGAL and cytokine levels were higher in patients with compartment syndrome. Biomarkers evaluated in this study may be used in the future as predictors of I/R injury severity and its possible evolution towards post‐reperfusion syndrome.

Introduction

Acute limb ischaemia (ALI) is a common and often severe clinical event 1. Its main causes are embolisation, in situ arterial thrombosis, acute graft occlusion 2, 3 or very infrequent complications of arterovenous fistula for haemodialysis 4.

ALI is related to a high frequency of mortality, particularly when the treatment is delayed 5, 6, 7.

Commonly, the first line of treatment is the reestablishment of blood flow, since it could be related with the development of several complications, for example, ischaemia reperfusion (I/R) syndrome 8, 9.

I/R syndrome has essentially two components: (i) local – activation and adhesion of polymorphonuclear cells (PNMs) and platelets, release of inflammatory mediators, that is, tumour necrosis factor‐alfa (TNF‐α), interleukin‐1 beta (IL‐1 β) and interleukin‐6 (IL‐6), activation of matrix metalloproteinases (MMPs), generation of free radical oxygen, cell death and increased tissue pressure with capillary occlusion (compartment syndrome) and increased ischaemic lesion; (ii) systemic – called post‐reperfusion syndrome (PRS), that is, secondary failure of organs and tissues related to the progression of the reperfusion injury 10, 11, 12.

MMPs are proteolytic enzymes that degrade the extracellular matrix (ECM) and play a role in both vascular 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 and non‐vascular diseases. Furthermore, it has been documented that MMPs are significantly involved in I/R injury in the lung, heart, brain and kidney 26. Neutrophil gelatinase‐associated lipocalin (NGAL) is an acute‐phase protein and its expression is upregulated under diverse conditions; it also cooperates with MMPs in degrading ECM 15, 16, 17, 18, 19, 23.

The aim of this article was to evaluate the role of MMPs, NGAL and inflammatory cytokines in patients with ALI, and their correlation with I/R syndrome and the main biochemical parameters.

Materials and methods

Study design

We conducted an open label, multicentric, parallel group study, between January 2010 and December 2013. This study was approved by the Institutional Review Board‐Independent Ethics Committee (IRB‐IEC). Before the beginning of the study, all participants were informed about the aim, procedures, risks and benefits of the study and a written informed consent was obtained from them.

Patients

Patients with acute embolism or thrombosis of the distal artery presenting with a diagnosis of acute ischaemia of the lower limbs were enrolled in this study.

Signs of an acute arterial occlusion were considered: slightly pale limbs; absence of proximal or distal pulses on affected limbs examined through Doppler ultrasound; decrease in active and passive movements of the limbs as well as peripheral reflexes; and impairment of sensation to tactile, thermal and nociceptive stimuli. Morphological graduation of the lesions was performed non‐invasively through duplex ultrasound and invasively during angiography.

ALI was evaluated in accordance with the Rutherford classification 27.

Enrolled patients were divided into two groups: group I – patients who required fasciotomy, and group II – patients who did not require fasciotomy, according to the onset of the compartment syndrome. We defined early compartment syndrome (ECS) as the onset of compartment syndrome within 2–6 hours after the revascularisation procedure and late compartment syndrome (LCS) as the onset of compartment syndrome over 12 hours after the revascularisation procedure 27, 28.

Fasciotomy has been advocated in patients with compromised blood flow for 4–6 hours, prolonged hypotension, combined arterial and venous injuries, distal limb swelling and absolute compartment pressures of 30 mmHg or higher, 40–55 mmHg, or 55 mmHg or higher 28.

Patients undergoing endovascular treatment with local thrombolysis as well as patients who had undergone unsuccessful surgical treatment resulting in amputation of an extremity were excluded. Patients without the possibility of adequate surgical revascularisation, resulting in limb loss, were excluded from this study because of failing evidence of reperfusion marker levels.

MMPs and NGAL evaluation in plasma and tissues

Blood samples were collected at the time of admission (T = 0), 6 hours (T = 1), 24 hours (T = 2), 3 days (T = 3) and 1 week (T = 4) after the revascularisation. Moreover, tissue biopsies were taken at the time of fasciotomy.

Samples were frozen and then analysed using enzyme‐linked immunosorbent assay (ELISA) test and Western Blotting to evaluate both MMPs and NGAL levels in plasma and tissues, respectively.

ELISA test

The ELISA test was performed as previously reported 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25. A commercially available sandwich ELISA kit with MMP‐1, ‐2, ‐3 and ‐9 monoclonal antibodies (Biotrak Human ELISA System, Amersham Pharmacia Biotech, Buckinghamshire, UK) was used to determine plasma MMP levels. The soluble concentrations of NGAL were quantified using a commercial ELISA kit (Bioporto diagnostics, Hellerup, Denmark), in agreement with the manufacturer's instructions. For both MMPs and NGAL, results were evaluated relative to a control group of healthy patients.

Western blot evaluation

During fasciotomy, tissues were taken and lysed in 2 ml of tissue protein extraction reagent (25 mM Bicine, 150 mM sodium chloride pH 7·6; Thermo Scientific, Waltham, MA). Immunoblotting was performed using anti MMP‐1, ‐2, ‐3, ‐9 and anti‐NGAL monoclonal antibodies as previously described 13, 21. All experiments were performed in triplicate. The pectineus muscle and the anterior tibial muscle (injured samples) were the sites of intraoperative biopsies.

Cytokine analysis

An ELISA (Human ELISA System, Amersham Pharmacia Biotech) kit was used to determine the plasma concentrations of IL‐1, IL‐6, IL‐8 and TNF‐α, as previously described 12, 17, 29, 30, 31.

Biochemical evaluation

Serum concentrations of creatine kinase (CK), lactate dehydrogenase (LDH), alanine aminotransferase (ALT), aspartate aminotransferase (AST), potassium and calcium ions (K⁺ and Ca⁺⁺), blood urea nitrogen (BUN), creatinine, c‐reactive protein (CRP), serum amylase and lipase, fibrinogen and albuminuria, and activated partial thromboplastin time (aPTT) were evaluated in all patients through the period of study (from T = 0 to T = 4).

Statistical analysis

To consider a difference in the response of matched pairs distributed with a standard deviation of 5 and a mean response of 2, we needed to study 55 pairs of subjects so as to be reject the null hypothesis that this response difference was zero with probability (power >83%, alpha 0·05, two‐tailed).

The mean, standard deviation and range were reported for the continuous variables, whereas the counts described the categorical variables. Data were checked for normality using the Kolmogorove–Smirnov test. The Student's t‐test and the χ 2 test were used when appropriate to test the significance of the differences. The analysis of variance (ANOVA) was used to evaluate the differences between multiple means. Once it had been determined that differences existed, a Bonferroni test was used to determine which means differed. Pearson's test was used to evaluate the correlation between biochemical markers and plasma levels of MMPs and of NGAL. All data were expressed as mean ± standard error of the mean (SEM). A P‐value of less than 0·05 was considered statistically significant. The SPSS (SPSS Inc., Chicago, IL) and G*Power (Institut für Experimentelle Psychologie, Heinrich Heine Universität, Dusseldorf, Germany) software were used for the statistical analyses.

Results

For this study, we enrolled 142 patients (96 men and 46 women, mean age 70·3 ± 7·1 years) with clinical signs of an acute arterial occlusion. The patients' characteristics are shown in Table 1.

Table 1.

Patients' characteristics

	Patients with fasciotomy	Patients without fasciotomy
	(n = 7)	(n = 135)	P value
Sex (M/F)	5/2	91/44	0·824
Comorbidity
Hypertension	4	98	0·735
Smoking	5	89	0·764
Diabetes mellitus	3	71	0·615
Atrial fibrillation	4	76	0·965
Ischaemic heart disease	5	82	0·571
Cerebrovascular disease	3	62	0·874
Anaemia	2	61	0·388
Renal insufficiency	5	96	0·986
Hepatic insufficiency	4	53	0·346
Malignancy	2	40	0·952
Dyslipidemia	6	101	0·514
Hyperuricemia	5	92	0·855
Obesity	6	64	0·048
CLI	4	49	0·266
Medication
Ace inhibitors	4	86	0·725
B‐blockers	5	83	0·597
Vasodilators	5	69	0·294
Lipid‐lowering agents	6	77	0·133
Thrombocyte aggregation inhibitors	7	90	0·064
Warfarin	3	62	0·874
Cause of occlusion
Embolism
Aortoiliac embolism	2	53	0·571
Cardiac embolism	3	57	0·973
Thrombosis
Occluded venous graft	0	6	0·568
Occluded synthetic graft	0	9	0·480
Occluded stent	1	6	0·240
Peripheral aneurysm	1	4	0·113
Distribution by severity
Rutherford I	0	9	0·480
Rutherford IIa	2	103	0·005
Rutherford IIb	5	23	0·0004
Rutherford III	0	0	—

CLI, chronic limb ischaemia; F, female; M, male.

ALI was most commonly related to atrial fibrillation (82 patients). Patent foramen ovale with contralateral embolism was found in nine cases. The time of limb ischaemia before surgical treatment was 2–48 hours. Time of limb ischaemia was not a significant predictor of clinically relevant compartment syndrome. The patients' history revealed that 53 of them suffered from chronic limb ischaemia (CLI): 19 in Fontaine IIa and 21 in Fontaine IIb, and 13 in Fontaine III.

The 142 patients were divided into two groups:

• Group I: patients (with compartment syndrome) requiring fasciotomy (n = 7; 4·93%).

• Group II: patients (without compartment syndrome) not requiring fasciotomy (n = 145; 95·07%).

Moreover, 70 healthy volunteers were enrolled in this study and represented the control group (group III).

The laboratory findings documented higher levels of all evaluated biochemical values in group I and group II relative to group III (P < 0·01; data not shown); moreover, the laboratory results were higher in group I relative to group II (P < 0·01; Table 2).

Table 2.

Laboratory findings recorded in enrolled patients after reperfusion. Results represent mean ± standard deviation

	Mean serum Levels after reperfusion
Laboratory measurements	Group I	Group II	P value
Creatine kinase (CK) (nr 24–195 U/l)	103 765 ± 1135	75 321 ± 932	<0·01
Lactate dehydrogenase (LDH) (nr 110–120 U/l)	29 004 ± 850	21 223 ± 784	<0·01
Alanine aminotransferase (nr 5–45 U/L)	8015 ± 335	5766 ± 178	<0·01
Aspartate aminotransferase (nr 5–45 U/L)	6280 ± 548	4399 ± 224	<0·01
Potassium ions (nr 3·5–4·5 meq/l)	5·3 ± 0·9	4·6 ± 0·7	<0·01
Calcium ions (nr 9–10 mg/dl)	12 ± 1·3	9·7 ± 1·5	<0·01
Blood urea nitrogen (nr 2·5–7·1 mmol/l)	9·3 ± 1·5	7·8 ± 1·3	<0·05
Creatinine (nr 60–110 μM/l)	254 ± 18·3	151 ± 15·2	<0·01
C‐reactive protein (CRP) (nr)	5·9 ± 0·8	3·5 ± 0·35	<0·01
Serum amylase (nr 30–100 U/l)	88 ± 11	64 ± 13	<0·01
Serum lipase (nr 10–60 U/l)	76 ± 14	46 ± 15	<0·01
Albuminuria (nr < 30 mg of albumin per 24 hours)	673 ± 42	294 ± 11	<0·01
Fibrinogen (nr 200–400 mg/dl)	989 ± 84	632 ± 55	<0·01
Activated partial thromboplastin time (aPTT) (nr 30–40 seconds)	56 ± 7·6	42 ± 7·2	<0·01

nr, normal range.

In group I, five of seven patients (71·43%) manifested an ECS and were treated with fasciotomy within 6 hours; of these, one patient (20·00%) died of multi‐organ failure; two of the seven patients (28·57%) showed LCS and were treated with fasciotomy within 24–36 hours; 1 patient (50·00%) died during the follow‐up period (Table 1).

The ELISA test revealed that plasma levels of MMP‐1, ‐2, ‐3 and ‐9 were higher in patients subjected to fasciotomy (group I) relative to patients not subjected to it (group II) (P < 0·01, Table 3). The Western blot evaluation also revealed an increased expression of active MMP‐1, ‐2, ‐3, ‐9 and NGAL in tissues of group I patients relative to patients enrolled in group II (P < 0·01, data not shown)

Table 3.

MMP plasma values obtained using ELISA test in patients subjected to fasciotomy (group I), in patients not subjected to fasciotomy (group II) and in control patients (group III). Results represent the mean ± standard deviation of values detected in enrolled patients

	At Time of Admission
	MMP1	MMP2	MMP3	MMP9	NGAL
Group I	5·2 ± 0·4	1025 ± 92·3	17·2 ± 2·1	425 ± 41·2	195 ± 33·4
Group II	4·1 ± 0·3	750 ± 58·7	11·5 ± 2·3	315 ± 32·5	145 ± 25·7
Group III	2·9 ± 0·7	478 ± 19·8	6·9 ± 1·1	54 ± 11·5	32 ± 12·3
	24–36 hours
Group I	4·4 ± 0·2	755 ± 85·1	12·1 ± 1·9	220 ± 36·3	91 ± 21·2
Group II	3·3 ± 0·4	630 ± 44·8	9·7 ± 1·4	132 ± 22·2	73 ± 19·4
Group III	2·8 ± 0·9	475 ± 19·5	6·8 ± 1·0	56·2 ± 12·3	34·8 ± 12·7
	1 week
Group I	2·9 ± 0·2	480 ± 20·3	7·2 ± 1·7	56·3 ± 12·1	34·9 ± 13·0
Group II	2·8 ± 0·6	478 ± 20·1	6·8 ± 0·8	55·7 ± 12·2	35·3 ± 12·6
Group III	2·8 ± 0·3	476 ± 19·7	6·9 ± 1·2	55·1 ± 12·6	33·1 ± 12·4

ELISA, enzyme‐linked immunosorbent assay; MMP, matrix metalloproteinases; NGAL, neutrophil gelatinase‐associated lipocalin.

Both plasma and tissue values of MMPs and NGAL as well as biochemical values decreased in a time‐dependent pattern after fasciotomy (see Table 3), with an initial decrease after 24–36 hours and a normalisation after 1 week; in patients with LCS, biochemical values decreased initially after 36 hours. Finally, using the Pearson's test we documented a significant correlation between MMP levels and biochemical parameters (P < 0·01).

Discussion

In this study, we evaluated the role of MMPs and inflammatory cytokines in patients with I/R syndrome of the lower limbs and their correlation with the main biochemical parameters.

Several studies have shown that I/R injury is related to endothelial cells (ECs) 32 and ECM activation: ECs activate both complement and coagulation systems and induce the recruitment of proinflammatory cells, which progressively induce the no‐reflow phenomenon 33. A dysfunction of both sodium‐potassium ATPase and calcium‐sodium exchanger is also involved; elevated levels of free calcium interact with actin, myosin and cellular proteases, resulting in skeletal muscle fibre necrosis. These events are amplified by reperfusion 34. Local complications of I/R injury may also result in amputation or severe dysfunction in the salvaged limbs 34. I/R injury of an extensive muscle tissue mass and the sensitive vascular tissues and endothelium often leads to systemic complications, a phenomenon called PRS 35. I/R injury is characterised by massive oedema often requiring fasciotomy, which increases fluid requirements and may cause shock; myoglobin, potassium, lactate and microthrombi release from the damaged skeletal muscle into the systemic circulation can cause renal failure, arrhythmias and, eventually, death 36.

In agreement with these data, in our article we reported that three patients died after fasciotomy, documenting that fasciotomy is usually performed in patients with impaired clinical conditions.

Several studies have shown a ‘common inflammatory pathway’ between lower limb injuries, such as I/R syndrome, and systemic organs: mediators of the local response to IR injury can also have systemic effects and the organs become the target tissues of a systemic inflammatory response process. Acute lung injury, myocardial dysfunction, acute renal failure, hepatic failure and pancreatitis contribute significantly to the morbidity and mortality associated with I/R syndrome 36.

Our experience confirms the central role of inflammation in I/R syndrome. We have shown the association between the main inflammatory cytokines and MMPs. In our study, patients with arterial thrombosis showed baseline elevated levels of MMP‐1, ‐2, ‐3, ‐9 and NGAL at the time of admission, compared to healthy volunteers; this is attributable to a pre‐existent chronic inflammatory vascular condition, as previously described 14, 15, 16, 17, 18, 19, 20, 21, 37. Moreover, in patients treated with fasciotomy because of compartment syndrome, higher levels of MMP‐1, ‐2, ‐3, ‐9 and NGAL were detected.

In accordance with the literature data 38, 39, 40, 41, our results document the association between inflammation and the main biochemical functional parameters.

Furthermore, we have highlighted the significant association between high biochemical parameter levels, which indicate damaged organs, and MMP‐1, ‐2, ‐3 and ‐9; another important aspect was that the biochemical parameters were much higher in patients who suffered the most serious complications of I/R syndrome, such as respiratory, renal and heart failure, than in patients who did not develop severe complications, and their levels of MMP‐9 were more elevated and associated with NGAL expression. NGAL can be considered as a parameter associated with an increased severity of the disease.

I/R injury is a potentially devastating condition associated with many diseases treated by surgery and plays a critical role in limb salvage procedures and in patients' functional outcomes. The authors have postulated that pharmacologic modulation of inflammatory cytokines is a promising strategy to prevent I/R syndrome 29. Our study focalised the pivotal role of MMPs: MMPs and NGAL could in the future be used as biological markers of I/R injury severity and its possible evolution towards more severe pathological stages such as PRS.