Tissue Inhibitors of Metalloproteinase 1 (TIMP-1) and 3 (TIMP-3) as New Markers of Acute Kidney Injury After Massive Burns

Abstract

Background

Acute kidney injury (AKI) is a common and serious complication after massive burn injury. One of the postulated etiologies is destruction of the extracellular matrix of nephrons, caused by a local imbalance between matrix metalloproteinases (MMPs) and specific inhibitors. The aim of this study was to analyze the dynamics of tissue inhibitors of metalloproteinases (TIMPs) during the first 5 days after massive thermal injury and the relationship with the risk of AKI.

Material/Methods

Thirty-three adults (22 men, 11 women) with severe burns were enrolled in the study. The values of TIMPs 1 to 4 were measured in blood serum and urine using the multiplex Luminex system. The associations between TIMPs and the risk of AKI were analyzed by using the generalized linear mixed models for repeated measurements.

Results

Significant changes in serum and urine activities of TIMPs were confirmed, especially during the first 2 days after burn injury. Almost half of patients presented renal problems during the study. Significant differences between values of TIMPs in AKI and non-AKI status were also observed. However, a significant relationship between concentration of TIMPs and risk of AKI was confirmed only for urine TIMP-1 and serum TIMP-3.

Conclusions

The evaluation of TIMPs in the early stage after burn injury has potential benefits. The important roles of urine TIMP-1 and serum TIMP-3, as novel markers of the risk of AKI development, were confirmed. Other parameters require further analysis.

Introduction

Acute kidney injury (AKI) is one of the components of burn injury. This frequent and serious complication of massive thermal trauma typically occurs during the first few days after a burn. AKI worsens the clinical course and patient survival. Diagnostic methods for this complication are currently insufficient, and the factors responsible for its evolution are not entirely understood [1].

A possible cause of AKI is the increase of activity of pro-inflammatory factors inside the renal parenchyma, which are released from thermally injured tissue [2]. Cytokines released into the bloodstream can damage the extracellular matrix (ECM) inside the nephrons by inducing specific proteolytic enzymes, for example, matrix metalloproteinases (MMPs). Lysis of support proteins and intracellular fibers and disintegration of renal interstitium structure occur. Vascular endothelial alterations cause the collapse of the vascular loop and acute ischemia among functional nephrons, ultimately leading to acute renal failure [3–5].

In response to severe inflammation and renal parenchyma ischemia, biologically active factors are locally induced to reduce destruction of nephrons and initiate repair processes. However, a consequence of rapid perfusion restoration is the risk of releasing accumulated toxic factors and developing acute ischemia-reperfusion injury in renal parenchyma and tubular damage [6].

MMPs are zinc atom-containing proteolytic enzymes, whose primary function is participating in the reconstruction and degradation processes of ECM elements [7]. Increased activity of MMPs is often observed during AKI and can be a poor prognostic factor for the recovery of kidney physiological functions, potentially worsening patient outcomes [8–10]. The precise impact of MMPs on the function of nephrons during AKI has not yet been sufficiently investigated. Researchers do not agree on the role of MMPs and suggest they may play a part in both renal parenchyma injury and protection and repair processes [11–13].

Activity of MMPs is strictly controlled by specific tissue inhibitors of metalloproteinases (TIMPs), which prevent their excessive accumulation in the ECM and disruption of its structure. In normal physiological conditions, MMPs and TIMPs are mutually inhibited in a negative feedback mechanism that maintains ECM reconstruction in a balance between absorption and synthesis [14]. So far, 4 types of TIMPs have been identified, TIMP-1-4, which regulate the activity of various MMPs [15].

AKI and burn injury can modify the serum and urine concentrations of TIMPs [16]. The etiology of this phenomenon is not well understood, but it is likely related to the protection and repair processes within the structures of injured organs. This can be caused by the inhibition of the destructive impact of secreted endopeptidases, blocking of active MMP forms, and restoration of the physiological equilibrium between the activity of MMPs and TIMPs [17,18].

The aim of this study was to evaluate the dynamics of the activities of TIMPs in blood serum and urine in patients during the first days after massive thermal injury and their impact on the risk of early phase of AKI occurrence.

Material and Methods

Study Group

In this prospective project, the study group was recruited from adult patients hospitalized in a specialist burn unit. The basic criteria for inclusion in the study group was an isolated massive thermal injury of second or third degree covering at least 30% of the total body surface area (TBSA) or at least 20% TBSA with an associated respiratory tract burn, which occurred no more than 72 h before hospitalization. Patients with other causes of burn injury, such as chemical burn and sunburn, mixed injuries connected with crush accidents, such as after explosions, and multi-organ traumas, such as broken bones or craniocerebral injuries, were also excluded. Histories of serious chronic renal, pulmonary, or cardiac dysfunction and active infections were also disqualifying criteria. According to the project protocol, patients were initially recruited during their primary hospitalization at the emergency ward and then evaluated by project team members in the participating clinical department. At the time of hospitalization, vital signs, level of consciousness using the Glasgow scale, and the severity of the burn, using the Abbreviated Burn Severity Index, were assessed.

Following the directions of the project, a unified treatment with high-volume fluid resuscitation was introduced at the time of hospital admission. During hospitalization, each patient was treated in accordance with current standards [19]. Fluid balance and volume status were strictly controlled throughout the observation period using objective diagnostic techniques, such as central venous pressure measurement and hematocrit range. In each patient, vital signs were evaluated, with special consideration given to fluid balance and daily urine collection.

AKI Diagnostics

In accordance with the objectives of the project, the following renal function parameters were estimated: blood serum urea (sUrea) and creatinine (sCr) concentrations, estimated glomerular filtration rate (eGFR), and hourly and daily diuresis. The diagnosis of AKI was based on the Acute Kidney Injury Network criteria and Risk, Injury, Failure, Loss of function, End-stage kidney disease (RIFLE) classification. In each patient, the current AKI status of the day was computed in the morning hours, along with other biochemical parameters. AKI-positive status was assigned to the appropriate RIFLE stages. The observation time was the first 5 days of hospitalization.

Determination of TIMPs

To implement the project assumptions, TIMP-1-4 values in blood serum (sTIMP-1-4) and TIMP-1-2 values in urine (uTIMP-1-2) were determined in the first 5 days of hospitalization in the morning hours. Serum and urine were prepared by centrifugation and stored at −80°C until being analyzed in a single batch.

Collection and Storage of Serum Aamples

Serum separator tubes were used and allowed blood samples to clot for 30 min at room temperature before centrifuging them for 15 min at 1000×g. Serum was removed immediately, and aliquots were stored at −80°C.

Collection and Storage of Urine Samples

The urine (mid-stream) samples were aseptically collected, voided directly into a sterile container, and centrifuged to remove particulate matter. Aliquots were stored at −80°C.

Description of the Methods used for TIMP Determination

On the day of the assay, all previously frozen serum samples were centrifuged at 16 000×g for 4 min immediately prior to use and diluted 50 times with test calibrator diluents. The urine samples were diluted 5 times with the same calibrator. The concentrations of TIMP-1-4 in serum and TIMP-1-2 in urine were measured in duplication using the Human TIMP-(1-4) Magnetic Luminex Performance 4-plex kit LKTM003 (R&D Systems, Inc, Minneapolis, MN, USA), which was designed for the simultaneous quantitative determination of the concentrations of multiple TIMPs in a single sample in cell culture supernates, serum, plasma, saliva, urine, and human milk. The Magnetic Luminex Performance 4-plex Kit LKTM003 assay was calibrated against highly purified recombinant human TIMPs produced at R&D Systems. Bio-Plex Manager 6.1 (Bio-Rad Lab Inc, Hercules, CA, USA) was used for the creation of 5-log standard curves, curve fitting, and calculation of concentrations for all analyses. Determinations of biological samples were performed in a certified hospital laboratory.

Reference Group

The reference group consisted of 10 healthy volunteers (5 men, 5 women) with an average age of 50.6 (±9.6) years and a negative history of kidney and other serious diseases. Concentrations of serum TIMP-1-4 and urine TIMP-1-2 were measured in this group as in the study group, but only once.

Bioethical Transparency

This study was approved by the authorized Bioethical Committee at the Military Institute of Medicine – National Research Institute (approval No. 36/WIM/2015). All performed techniques followed the currently applicable regulations of conducting research projects. Patients recruited to the study group and volunteers from the control group provided voluntary written consent for participation in this research. They had complete insight into the project objectives, were informed about its progress, and had the opportunity to withdraw at any stage of the investigation.

Statistical Analysis

For crude data, the results were presented as medians (range), due to the non-symmetrical distribution. For comparisons of the studied parameters on each day with the reference values, the one-sample Wilcoxon test was applied. For comparisons of the studied parameters on each day between AKI and non-AKI status, the exact Mann-Whitney U test was used. To investigate the relationship between studied parameters, time from injury, and AKI status, generalized linear mixed models (GLMM) for repeated measurements were estimated. The interaction term was taken into account, to reflect the possible differences in the course of studied parameters in relation to AKI. A gamma probability distribution of the error term, log link function, and first-order autoregressive covariance structure were assumed based on the Akaike information criterion. Asymmetry of distributions and the need for comparability between estimated and crude data were taken into account by using the gamma probability. The effect of risk factors on hospitalized patients was measured by mean values with 95% confidence intervals (CI) as a function of the number of days from injury and AKI status.

A P value <0.05 was considered to be statistically significant. Statistical analyses were performed in IBM SPSS version 25 and SAS/STAT version 15.2.

Results

Characteristics of the Study Group

The study included 33 adult patients (22 men, 11 women) with an average age of 48 (±18.58) years, and with BMI and TBSA values in the upper limits of the norm for the Polish population. Table 1 presents the general characteristics of the study group.

Table 1.

Characteristics of the studied group of patients.

Characteristics	N=33
Sex (M/F)	22/11
Age (years), mean (±SD)	48 (±18.58)
BMI (kg/m2), mean (±SD)	25.77 (±5.42)
TBSA (m2), mean (±SD)	1.89 (±0.25)
Burned TBSA (%), mean (±SD)	42.21 (±18.14)
3rd degree burns, n (%)	22 (66.66)
TBSA with 3rd degree burns (m2), mean(±SD)	12.36 (±7.27)
Respiratory tract burns, n (%)	23 (69.7)
Consciousness disturbances (GCS), n (%)
None or mild (13–15 pts)	8 (24,24)
Moderate (9–12 pts)	0
Severe or intubation (<8 pts)	25 (75,76)
ABSI score (pts), mean (±SD)	5.82 (±2.2)
APACHE II score (pts), mean (±SD)	15.27 (±8.13)
Patients with AKI-positive status, n (%)	15 (45.5)
AKI-positive status in RIFLE classification
Risk, n (% active AKI-positive)	14 (93.33)
Injury, n (% AKI-positive)	13 (86.67)
Failure, n (% AKI-positive)	10 (66.67)
Requirement of renal replacement therapy, n (% AKI-positive)	5 (33,33)
Administration of pressor amines, n (%)	22 (66.67)
Death/survival, n (%)	16 (48.48)/17 (51.52)

SD – standard deviation; BMI – body mass index; TBSA – total body surface area; GCS – Glasgow coma scale; ABSI – Abbreviated Burn Severity Index; AKI – acute kidney injury; RIFLE – Risk, Injury, Failure, Loss-of-function, End-stage classification; n – number of patients.

TIMP-1-4 Concentration Values and AKI Risk

The observed individual median concentration values of TIMP-1-4 are presented in Table 2, depending on AKI status (AKI and non-AKI), number of days of observation that had elapsed since the injury, and their relationship with the reference values.

Table 2.

TIMP-1-4 concentration values (ng/mL) obtained during the observation period.

	Days from injury	AKI-positive status			AKI-negative status			AKI vs Non-AKI status p-value	Reference value (min–max)
		n	Median (min–max)	P value*	n	Median (min–max)	P value*
sTIMP-1	1	7	451.99 (227.29–570.99)	0.016	1	132.75	–	–	145.56 (117.52–185.4)
	2	6	433.95 (216.66–594.06)	0.031	10	209.67 (125.61–346.79)	0.020	0.005
	3	5	238.75 (195.08–407.84)	0.063	19	188.78 (124.36–664.34)	<0.001	0.008
	4	1	224.45	–	14	186.57 (148.69–358.54)	<0.001	–
	5	1	197.85	–	7	169.96 (160.39–193.88)	0.016	–
uTIMP-1	1	7	443.95 (9.18–585.14)	0.016	1	9.78	–	–	6.47 (0.49–13.54)
	2	6	31.24 (3.69–162.17)	0.094	10	23.68 (2.87–138.7)	0.020	0.792
	3	7	15.1 (8.76–86.19)	0.016	19	5.38 (1.46–86.26)	0.760	0.018
	4	3	22.87 (17.27–30.72)	0.250	15	11.7 (2.11–54.11)	0.010	0.250
	5	3	29.05 (5.68–38.90)	0.500	7	15.94 (3.35–47.32)	0.109	0.667
sTIMP-2	1	7	100.61 (93.47–126.72)	0.375	1	81.17	–	–	113.16 (102.67–134.34)
	2	6	109.40 (82.65–117.87)	0.313	10	101.69 (75.17–118.41)	0.065	0.492
	3	6	99.44 (62.99–122.25)	0.156	19	91.72 (69.96–156.35)	0.001	0.780
	4	2	93.66 (84.25–103.07)	0.500	14	94.45 (74.31–118.95)	0.001	0.600
	5	2	87.94 (75.03–100.84)	0.500	7	85.50 (74.6–105.51)	0.016	1.000
uTIMP-2	1	7	95.73 (17.63–936.83)	0.031	1	58.28	–	–	29.97 (0.38–76.66)
	2	6	40.40 (14.03–150.33)	0.313	10	48.43 (14.51–100.27)	0.037	0.875
	3	7	51.32 (19.32–2844.14)	0.109	19	24.83 (9.07–1719.38)	0.623	0.063
	4	3	38.71 (34.39–88.45)	0.250	15	31.23 (6.13–61.32)	0.934	0.164
	5	3	39.54 (26.29–71.08)	0.500	7	32.11 (4.02–65.12)	0.934	0.667
sTIMP-3	1	7	50.40 (42.52–74.07)	0.016	1	23.63	–	–	33.78 (23.63–83.03)
	2	6	54.03 (33.78–64.49)	0.063	10	40.40 (17.59–54.1)	0.172	0.048
	3	7	46.55 (23.63–77.11)	0.203	19	38.28 (23.63–61.13)	0.182	0.412
	4	3	46.55 (46.55–67.76)	0.250	14	40.40 (28.94–61.13)	0.041	0.115
	5	3	50.4 (46.55–57.67)	0.250	7	33.78 (28.94–38.28)	0.625	0.008
sTIMP-4	1	7	3.38 (1.82–5.51)	0.031	1	1.24	–	–	1.91 (1.33–2.29)
	2	6	2.86 (2.14–4.73)	0.031	10	1.75 (1.24–3.1)	0.264	0.004
	3	7	2.33 (1.47–7.72)	0.109	19	1.72 (1.01–4.51)	0.852	0.065
	4	3	3.32 (2.42–3.81)	0.250	14	2.03 (1.33–3.76)	0.542	0.068
	5	3	2.88 (1.82–3.24)	0.500	7	1.98 (8.23–3.28)	0.938	0.383

(^*) comparison with the reference value. TIMP 1–4 – tissue inhibitor of metalloproteinases 1–4; s – blood serum; u, urine; AKI – acute kidney injury; n – number of observations.

The mean values of TIMP-1-4 in blood serum (sTIMP-1-4) and TIMP-1-2 in urine (uTIMP-1-2) estimated using GLMM are presented in Table 3, depending on the AKI status and the number of days since injury.

Table 3.

Estimated means of TIMP-1-4 parameter concentrations with acute kidney injury status and the number of days from injury (GLMM model).

	Overall effects P values	Days from injury	Mean (95% CI)		Ratio of Means	P value
			AKI status (ng/mL)	Non-AKI status (ng/mL)
sTIMP1*	AKI status: P=0.216 Days from injury: P=0.100 Interaction term: P=0.394	1	351.32 (274.26–450.03)	214.52 (123.86–371.53)	1.64	0.091
		2	315.97 (242.69–411.38)	249.97 (203.53–307.02)	1.26	0.105
		3	229.02 (171.26–306.27)	230.05 (192.69–274.65)	1.00	0.977
		4	217.38 (122.85–384.63)	224.22 (185.20–271.46)	0.97	0.916
		5	228.70 (132.45–394.89)	208.57 (163.15–266.64)	1.10	0.743
uTIMP1	AKI status: P= 0.037 Days from injury: P=0.268 Interaction term: P=0.230	1	205.29 (89.19–472.48)	13.14 (1.50–115.04)	15.63	0.021
		2	44.19 (18.08–108.03)	38.36 (19.17–76.76)	1.15	0.802
		3	34.20 (14.75–79.31)	14.61 (8.74–24.44)	2.34	0.088
		4	18.74 (5.32–65.96)	18.42 (10.39–32.68)	1.02	0.980
		5	24.31 (6.82–86.63)	18.59 (8.09–42.73)	1.31	0.725
sTIMP2**	AKI status: P=0.441 Days from injury: P=0.231 Interaction term: P=0.503	1	110.74 (98.62–124.34)	88.88 (66.20–119.33)	1.25	0.167
		2	108.53 (95.86–122.87)	99.72 (90.60–109.75)	1.09	0.271
		3	94.86 (83.55–107.72)	96.75 (89.92–104.10)	0.98	0.786
		4	90.72 (73.66–111.74)	97.95 (90.14–106.43)	0.93	0.492
		5	89.96 (72.8–111.18)	88.78 (79.07–99.67)	1.01	0.912
uTIMP2	AKI status: P=0.222 Days from injury: P=0.053 Interaction term: P=0.903	1	228.26 (54.82–950.53)	59.00 (1.36–2569.81)	3.87	0.506
		2	58.29 (12.49–272.09)	49.27 (14.94–162.51)	1.18	0.864
		3	448.88 (107.79–1869.34)	121.11 (50.95–287.90)	3.71	0.122
		4	52.63 (5.96–465.05)	31.10 (11.74–82.42)	1.69	0.662
		5	45.57 (5.16–402.70)	32.32 (7.76–134.60)	1.41	0.793
sTIMP3	AKI status: P<0.001 Days from injury: P=0.563 Interaction term: P=0.489	1	54.5 (44.84–66.26)	27.76 (16.70–46.17)	1.96	0.016
		2	53.96 (43.76–66.53)	38.65 (32.85–45.47)	1.40	0.013
		3	45.65 (37.49–55.59)	38.22 (33.89–43.11)	1.19	0.127
		4	50.97 (37.95–68.45)	41.89 (36.47–48.12)	1.22	0.232
		5	50.81 (37.73–68.43)	36.27 (29.84–44.09)	1.40	0.062
sTIMP4	AKI status: P=0.074 Days from injury: P=0.560 Interaction term: P=0.115	1	4.31 (3.18–5.83)	1.74 (0.85–3.54)	2.48	0.019
		2	3.20 (2.33–4.38)	2.12 (1.65–2.73)	1.51	0.028
		3	2.71 (1.97–3.71)	2.22 (1.81–2.74)	1.22	0.266
		4	2.02 (1.32–3.10)	2.44 (1.94–3.05)	0.83	0.416
		5	1.98 (1.26–3.10)	2.00 (1.48–2.72)	0.99	0.957

^*Patients 33 and 11 were deleted from the analysis for having extreme values (>3*IQR) in all points of observation.

^**Patient 33 was deleted from the analysis for having extreme values (>3*IQR) in all points of observation.

GLMM – generalized linear mixed models; TIMP 1–4 – tissue inhibitor of metalloproteinases 1–4; s – blood serum; u – urine; AKI – acute kidney injury.

Figures 1 to 4 show the estimated mean and 95% CI calculated in the multivariate GLMM model for TIMP-1-4 concentrations within the first 5 days after burn injury.

Figure 1.

Estimated means and 95% CI for serum (s) and urine (u) tissue inhibitor of metalloproteinase-1 (TIMP-1) concentrations during first 5 days since burn injury by acute kidney injury status (Microsoft Excel, Microsoft 365 MSO). (A) sTIMP-1; (B) uTIMP-1. Dotted line indicates the reference value.

Figure 2.

Estimated means and 95% CI for serum (s) and urine (u) tissue inhibitor of metalloproteinase-2 (TIMP-2) concentrations during first 5 days since burn injury by acute kidney injury status (Microsoft Excel, Microsoft 365 MSO). (A) sTIMP-2; (B) uTIMP-2. Dotted line indicates the reference value.

Figure 3.

Estimated means and 95% CI for serum tissue inhibitor of metalloproteinase-3 (sTIMP-3) concentrations during first 5 days since burn injury by acute kidney injury status (Microsoft Excel, Microsoft 365 MSO). Dotted line indicates the reference value.

Figure 4.

Estimated means and 95% CI for serum tissue inhibitor of metalloproteinase-4 (sTIMP-4) concentrations during first 5 days since burn injury by acute kidney injury status (Microsoft Excel, Microsoft 365 MSO). Dotted line indicates the reference value.

The average values of sTIMP-1 and uTIMP-1 were significantly higher than the reference values (Tables 2, 3; Figure 1A, 1B). Generally, sTIMP-1 values were higher than those of uTIMP-1, and the tendency to decrease was slower. In instances of positive AKI status, the dynamics of sTIMP-1 and uTIMP-1 value changes were approximately parallel. High amounts persisted during the first 1 to 2 days, and subsequently, there was a decrease toward values comparable with non-AKI status; however, in urine this process was faster.

No relationship between sTIMP-1 concentration and the number of elapsed days since the injury and AKI risk was demonstrated in the analysis for multiple measurements. However, a significant relationship between uTIMP-1 concentration and AKI occurrence (P=0.037) was shown. A significantly higher uTIMP-1 concentration on day 1 after injury in instances of AKI (P=0.021) was also confirmed (Table 3).

The average concentration of sTIMP-2 in the study group remained below the reference group values during the entire period of observation (Tables 2, 3; Figure 2A). In AKI status, sTIMP-2 tended to slowly decrease during all the days of observation. In instances of non-AKI status, the levels of sTIMP-2 were more stable, but no significant differences between the results obtained during AKI and non-AKI status were found. However, in the analysis of multiple measurements, no association between sTIMP-2 concentration and the number of days that had elapsed since injury and AKI occurrence was found (Table 3).

The dynamics of uTIMP-2 value changes throughout the time of observation differed from those of sTIMP-2 values. Alterations of uTIMP-2 concentration in AKI instances, as well as non-AKI, were irregular and were generally above reference values, especially in AKI-positive status (Tables 2, 3; Figure 2B). In the analysis of multiple measurements, no relationship between uTIMP-2 concentration and the number of days that had elapsed since injury and AKI occurrence was found (Table 3).

The average values of sTIMP-3 in the study group remained mostly above the reference values, especially in AKI-positive status. Significantly higher activity in instances of AKI-positive status than in AKI-negative status were also confirmed in the first days after injury (P=0.016 on day 1, P=0.013 on day 2; Tables 2, 3; Figure 3). Additionally, the relatively strong relationship between sTIMP-3 and AKI occurrence was confirmed in the analysis of multiple measurements (P<0.001), but no association between sTIMP-3 concentration and the number of days that had elapsed since injury was shown (Table 3).

The maintained high activity of sTIMP-4 in AKI-positive status in comparison to non-AKI status, and the values of reference concentrations in the first 4 days after injury were visualized in our observations. The average concentrations of sTIMP-4 in AKI and non-AKI status were similar on days 3 to 5, reaching values close to the average reference concentration. The results have a tendency toward a slight increase during the first 4 days in non-AKI status (Tables 2, 3; Figure 4).

The relationship between sTIMP-4 and AKI risk was not confirmed, but significantly higher concentrations of sTIMP-4 in the first days after injury in AKI instances (P=0.019 on day 1, P=0.028 on day 2) were confirmed in the analysis for multiple measurements (Table 3).

Discussion

Massive burn injuries are among the rarer causes of hospitalization. However, since they lead to many life-threatening complications, they are still a serious challenge for clinicians. Diagnostics and treatment of developing multi-organ disorders are the most important problems.

The present study included 33 adult patients, mostly in a bad health and requiring intensive care. The patients were massively burnt, with a median injured TBSA of 42.21% (±18.14%). Approximately two-thirds of patients presented with third-degree burns of skin integuments, with a median 12.36% (±7.27%) TBSA, and in 69.7% of patients, thermal injury of the respiratory tract was diagnosed. The health status of three-quarters of the patients was determined as serious or critical (Glasgow coma scale <8), and two-thirds of patients required vasopressor support during the observation period. The median Abbreviated Burn Severity Index value was 5.82 (±2.2) points, suggesting an average survival probability of around 80%. The mean APACHE II score of 15.27 (±8.13) points showed serious disturbances of the patients’ conditions.

AKI is a common and serious complication after massive burns. During our 5-day observation, almost half of patients presented problems with kidney function, and one-third of the AKI-positive patients required renal replacement therapy. Patients with AKI were in significantly worse condition and more often required amine pressors.

The pathogenesis of post-burn AKI is still unclear. One of the main causes of this process can be the impact of certain cytokines released from cells directly damaged by thermal energy. Bioactive substances reach the kidneys through the bloodstream and cause an escalation of inflammation inside kidney tissue. Severe inflammation leads to nephron dysfunction and metabolic disorders [20].

One of the potential AKI mechanisms is destruction of renal ECM, due to disruption of the enzyme secretion system: extracellular MMPs and their TIMPs [21].

The main function of MMPs is hydrolysis and removal of ECM components, mainly collagen, in the process of tissue resorption and reconstruction. Increased activity of MMPs is found within damaged nephrons. The exact role of MMPs in AKI mechanisms is not fully known. MMPs affect the ECM structure by damaging bonds between the cells of glomeruli, tubules, and intra-renal vessels endothelium, which are their support and are responsible for the adhesion of cells to each other. The anatomy of the renal tissue is loosened and destroyed, causing a dysfunction of active renal parenchyma [7]. Pathological changes within the endothelium support the collapse of intra-renal vessel loops, with simultaneous acute ischemia of nephrons and a fierce accumulation of toxic elements, such as oxygen free radicals, reactive forms of oxygen, and nitric oxide [22].

Some previous authors indicated and described particular elements of the renal ECM, whose damage was responsible for the AKI process. Hu et al highlighted the impact of MMP-9 and MMP-2 on the degradation of tight junctions bonding renal cells [23]. Covington et al confirmed the role of MMP-14 in the proteolysis of cadherin, an integral component of the ECM of renal tubules [24]. Caron et al described the role of MMP-9 in the destruction of occludin protein, which is responsible for maintaining bonds between the vessel endothelial cells in the renal cortex [25].

In response to increased MMP activity, surrounding cells and tissues react by secreting specific inhibitors (TIMPs) to the extracellular space. Four TIMPs (TIMP-1-4) have been identified so far. TIMP secretion can be regulated by cytokines and growth factors. While TIMP-2 expression is constitutive and ubiquitous, TIMP-1, TIMP-3, and TIMP-4 expression is inducible and tissue specific. The primary role of TIMPs is the inhibition of proteolytic activity of MMPs. They can bind not only to MMPs but also to proenzymes (proMMPs), so their activation process can also be controlled [21].

Only a few reports dealing with the link between TIMP activity and the risk of AKI are currently available. Potentially, TIMPs neutralizing MMPs may have a protective influence on the structures of nephrons by inhibiting the progression of ECM destruction and reducing the risk of renal parenchyma ischemia. Additionally, TIMPs, by regulating the dynamics of changes within ECM, can have a nephroprotective function by impacting the reperfusion process of already ischemic tissues. Slowing down the release process of accumulated toxic substances also potentially reduces the risk of ischemia-reperfusion injury [26].

In the present study, an original prospective paper, an attempt at innovative imaging and dynamic evaluation of simultaneous, parallel changes of all known types of TIMPs to date was completed. This was achieved by multiple evaluations in blood serum and/or in urine during specified time intervals from the moment of massive thermal injury. An in-depth analysis of dependencies between the obtained values of TIMPs and risk of AKI in this group of patients was conducted.

The study focused on the analysis of all known types of TIMPs, to make the study results clear and reliable. The mechanisms of the balance between MMPs and TIMPs are complicated and not completely clear. Each type of TIMP can influence several types of MMPs. On the other hand, the activity of each MMP can be changed by more than 1 variant of TIMP. Additionally, the impact can be modified depending on the local conditions and the tissue environment. The comparisons with MMP activities will be a subject of further studies because of their omnidirectional and complicate relationships. The main point of interest of the study was the systemic influence of TIMPs; therefore, the local activities of TIMPs (ie, in the burned tissues) were not measured.

Current guidelines on post-burn AKI diagnostic recommend measurement of urine output and blood serum laboratory tests (concentration of creatinine and urea). Renal biopsies are not a routine practice in such cases, especially at the early stages after thermal injury; therefore, they were not performed in the present study, and the histology material was not available for the analysis.

Tissue Inhibitor of Matrix Metalloproteinase 1

TIMP-1 is the first discovered and best-known protein of the TIMPs group, with a molecular weight of 30 kDa [27]. It can inhibit almost all types of MMPs [28]. In a series of publications, an increase in TIMP-1 value during AKI was confirmed. In the paper by Huang et al, a link between uTIMP-1 increase and risk of AKI development and a decline in survival of children in serious condition was demonstrated [29]. Bojic et al showed a positive association between the increase in sTIMP-1 and the sepsis-associated type of AKI [30]. Ozkan et al revealed an analogous association with sTIMP-1 and uTIMP-1 in patients with heterogenic etiology of AKI [31]. However, Wang et al confirmed a high value of sTIMP-1 increase for sepsis prediction, severe clinical course, and patient death, but they did not show a significant relationship with risk of AKI [32].

In the present study, significantly higher values of sTIMP-1 and uTIMP-1 in comparison to the reference values were confirmed. The values of sTIMP-1 were higher than those of uTIMP-1. The tendency for the largest changes in TIMP-1 in both serum and urine during first 24 and 48 h from thermal injury was confirmed. Longer continuation of TIMP-1 activity in blood serum than in urine was also found, which was consistent with the results cited above.

Previous reports, including those with the participation of burn patients, confirmed post-burn disorders of TIMP-1 activity in response to the systemic pro-inflammatory state. Nagy et al noted an increase in sTIMP-1 from day 2 until the end of a 5-day observation period and its weak positive association with patient survival [33]. In the study by Dasu et al, higher TIMP-1 activity in children with burns, but without significant dynamics during the several days of observation, was demonstrated [34]. Although the work of Hasback et al was based on a single sTIMP-1 measurement in each patient, the obtained results are grouped in terms of the hour/day intervals from the burn. The maximization of sTIMP-1 values during the first 48 h was confirmed in the cited analysis, which is consistent with our results. Additionally, a link between TIMP increase and severity of injury and poor survival prognosis of patients was shown [35]. Ulrich et al presented a significant sTIMP-1 rise during the first 3 days in patients with severe burns treated surgically [36]. Conversely, the report by Stanciu et al describes a slow increase of sTIMP-1 up to 7 days after injury [17].

In the present study, no relationship between sTIMP-1 and the number of days since injury and AKI risk was demonstrated. Only a significant relationship between uTIMP-1 concentration and AKI occurrence and higher uTIMP-1 concentration at the closest time after injury in instances of AKI was confirmed.

However, the relationship between TIMP-1 value alterations and risk of AKI development associated with burns was not analyzed in any of the cited studies. The absence of consistent results for measurements in serum and in urine is unclear and requires further investigation.

Tissue Inhibitor of Matrix Metalloproteinase 2

TIMP-2 is a 24 kDa protein. TIMP-2 is known to form high-affinity complexes with almost all types of MMPs. TIMP-2, unlike the other TIMPs, is not inducible by growth factors. A distinctive feature of TIMP-2 is its ability to selectively regulate MMP-2 activity. TIMP-2 is the only TIMP that can both inhibit and activate MMP. The effect of TIMP-2 on MMP-2 depends on its concentration values – low amounts stimulate MMP-2 expression and high amounts inhibit MMP-2 activity. Thus, depending on the biological situation, TIMP-2 can intensify or inhibit ECM degeneration and nephroprotection processes [37,38].

Additionally, TIMP-2 has the unique potential to directly block the proliferation of endothelial cells. The anti-proliferation mechanism is associated with an exceptionally strong ability to inhibit the cell cycle in the G1 phase. This is a protective function against DNA destruction in the case of sudden cell ischemia in critical situations. It can have a positive effect in the emergency situation of ischemia-reperfusion injury syndrome occurrence by restraining secondary tissue destruction after reperfusion. Similar activity is also presented by insulin-like growth factor binding protein 7 (IGFBP-7), and both cytokines are termed cell cycle arrest biomarkers [39].

Unfortunately, only a few papers are available that describe the prognostic sTIMP-2 value as a potential AKI marker. In the work by Wang et al, the ambiguous and dualistic role of sTIMP-2 in the AKI mechanism was indicated as being caused by simultaneously inhibiting MMPs (especially MMP-2) and promoting their synthesis in the gene transcription phase [40].

The most specific and characteristic features of sTIMP-2 during the entire period of observation were its values below the reference values. The sTIMP-2 values in AKI-positive status, similarly to that of sTIMP-1, tended to slowly decrease throughout the days of observation. In non-AKI instances, the sTIMP-2 levels presented a similar tendency but were more stable.

This is probably related to the unique ability of TIMP-2 to alternate the possibility of MMP-2 stimulation/inhibition, which is characterized by one of the strongest potentials for ECM remodeling. It is well known that higher TIMP-2 values (observed in the first days) inhibit MMP activity, protectively impacting ECM and decreasing the risk of destruction. The decrease in activity in the further part of observation can indicate a rising tendency of MMP-2 stimulation to initiate remodeling and remove damaged ECM elements, potentially reducing the risk of organ fibrosis. However, no association between sTIMP-2 concentration and the number of elapsed days since the injury and AKI occurrence was confirmed in the present study.

The association between increase of uTIMP-2 or uTIMP-2*IGFBP-7 product and risk of AKI development was described in recent publications. This applied to AKI with post-contrast etiology (Sun et al), septic etiology (Molinari et al, Kashanii et al), after sudden cardiac arrest (Adler et al), postoperative (Wang et al), and posttraumatic (Sakyi et al) etiologies [41–45]. However, studies presenting uTIMP-2 alone or together with uIGFBP-7 in burn patients are limited. Only a few reports showing the utility of these biomarkers in early prediction and diagnosis of AKI in high-risk patients with various injuries, including burns, are available [46].

The dynamics of uTIMP-2 differed from those of sTIMP-2 in the present study. The values of uTIMP-2 were more variable and above reference values, especially during the entire time of observation in AKI-positive status. The increase in levels to the maximal amounts on day 3 was noticeable. This is probably associated with the hypothesis that renal tubular cells are the main source of uTIMP-2 synthesis and secretion, as presented in the previously cited study by Kashani et al. Disintegration of interstitial cells in response to post-thermal damage of nephrons causes rapid TIMP-2 release in the renal tubular lumen and an increase in concentration in urine [39]. However, the link between the increase of uTIMP-2 and the risk of AKI shown in this publication was relatively weak and not statistically significant. Additionally, no relationship between uTIMP-2 and the number of elapsed days since injury and AKI occurrence was shown.

Tissue Inhibitor of Matrix Metalloproteinase 3

TIMP-3 is expressed in various human tissues. The molecular weight of TIMP-3 is about 24 kDa [47]. TIMP-3 is unique among the TIMPs because of its broad inhibitory range. It can deactivate almost all types of MMPs, including a number of ADAMs (a disintegrin and metalloproteases), and ADAMTS (ADAM with thrombospondin motifs) [48]. TIMP-3 can not only inhibit the activated form of MMPs but can also interrupt the promotion of pro-enzymes. TIMP-3 can also modify the angiogenesis process through suppression of vascular endothelial growth factor [49]. TIMP-3 is the only one from the TIMP group that does not exist in a soluble form, so it occurs only in the ECM and basal membranes [50].

The level of TIMP-3 is relatively high in healthy kidney tissue. Its decrease can be observed during tubulointerstitial fibrosis [51]. The link between decreased TIMP-3 concentrations and a greater risk of renal interstitial damage was shown in the study by Wang et al. However, the interpretation of these observations is still unclear. The authors emphasized the nephroprotective impact of TIMP-3, resulting from blocking MMP-2 synthesis and secretion, and subsequently rapid nephron fibrosis inhibition [44]. Daniels et al described the positive effect of TIMP-3 on the return to normal kidney function after AKI, and a reduction in demand for renal replacement therapy was demonstrated in their publication. The authors related this finding to the strong suppression of kidney inflammation and fibrosis [52]. Similarly, Zhou et al showed the influence of increased TIMP-3 activity on risk reduction of AKI with an ischemic mechanism. The authors justified this by the especially strong inhibition of MMP-2 and MMP-9 [53].

The values of sTIMP-3 in the study group remained mostly above the reference values in the present study. Instances of AKI-positive status were also significantly higher than that of AKI-negative status. This was likely related to the potent TIMP-3 action described above, which suppressed the impact of multiple MMPs. These enzymes are well-known strong stimulators of post-burn inflammation. Additionally, a relevant relationship between sTIMP-3 and AKI occurrence was shown, but with no association between sTIMP-3 and the number of days from injury. Unfortunately, no data dealing with the role of TIMP-3 in the course of burns were found in the available literature. However, judging by the results of the presented work and the references cited above, potential benefits of TIMP-3 evaluations in the diagnostics of post-burn AKI can be expected.

Tissue Inhibitor of Matrix Metalloproteinase 4

TIMP-4 is the last and weakest known member of the TIMP family. Its molecular weight is about 25.5 kDa. The TIMP-4 molecule shows the greatest similarity in structure to TIMP-2. The essential difference between TIMP-4 and TIMP-2 concerns the C-terminal domain sequence, determining the unique influence of MMP-2 activity. This is the cause of the similar bioactive properties of TIMP-4 and TIMP-2, but only within the scope of MMP inhibition [54].

In our observations, high activity of sTIMP-4 in AKI-positive status in comparison with non-AKI status and the values of reference concentrations were observed, especially just after the time of injury. Comparing the graphs of sTIMP-4 and sTIMP-2 (Figures 2A, 4), a specific analogy in the course of the concentration curves of AKI/non-AKI during the entire period of observation could be seen. However, the sTIMP-2 results reached values on average around 100 times higher than those of sTIMP-4. Additionally, the sTIMP-2 and sTIMP-4 result curves were positioned differently than those of the reference data. This was probably associated with the unique TIMP-2 feature, which was based on the induction capabilities of MMPs in low-concentration conditions. However, a significant relationship between sTIMP-4 and AKI risk was not confirmed.

Summary of TIMPs

A similar configuration could be seen in the distribution of all types of TIMP data. The AKI-positive agents generated the highest TIMP values. The non-AKI status values were localized mostly between AKI-positive and reference curves. The reference values were situated at the bottom of the charts. The only exception was the case of sTIMP-2, for which a mirror inversion could be seen; this was probably connected with the unique ability of TIMP-2 to not only inhibit but also to promote the MMP activities.

Although statistical significance was not confirmed in all cases and types of TIMP, the differences between values in AKI status could be clearly seen. The increased level of blood serum and urine TIMPs could be treated as a useful AKI marker after burn injury.

Also worth mentioning is the mutual relationship between the AKI-positive, AKI-negative, and reference values. The maximum discrepancy could be seen mostly in the first 48 h after the injury. In the subsequent days, the courses of the graph curves became closer to each other, nearly overlapping at the end of the observation time.

This confirms that the time of maximum TIMP storm occurs during the first 48 h after the burn injury. During this time, their potential power as AKI biomarkers seems to be the greatest. This is likely the best moment for intensive testing of these markers in the blood serum and urine to improve the AKI diagnostics. The levels of cut-off points for the increase of TIMPs should be carefully standardized. This can be a goal for future investigations. The greatest limitation is the lack of wide-ranging publications describing this important clinical problem.

Study Limitations

The small number of patients affected by massive thermal injuries during the project period did not allow us to prove the existence of all observed relationships. However, statistical power of 60% to 70% was obtained when testing the significance of the associations. Also, the different times from the moment of thermal injury to the first TIMP evaluation appeared to be a serious technical problem, which resulted in different time intervals for the first evaluation. To account for this fact, mixed longitudinal models were used. Finally, the activities of TIMP-3-4 were analyzed only in blood serum. None of the certified laboratory tests validated for measurements of these parameters in the urine were achievable at the time of implementation of the project.

Conclusions

TIMPs seem to play an important role during the early phase of AKI processes in patients with severe burns as indicators of damage to the nephrons’ ECM. Systemic evaluations of TIMPs, both in blood and in urine, have potential benefits and improve AKI diagnostics, being helpful in assessing the risk of its occurrence at an early stage. The important roles of uTIMP-1 and sTIMP-3, and their relationships with AKI were confirmed. They can be considered as the potential novel markers of AKI risk in the first days after thermal injury. Analyses of other possible associations of TIMPs require further studies with a larger cohort of patients, taking into account measurements of uTIMP-3 and uTIMP-4.