Pentraxin‐3: A new parameter in predicting the severity of diabetic foot infection?

Abstract

This study was undertaken to evaluate the diagnostic and prognostic values of pentraxin‐3 (PTX‐3) in patients with infected diabetic foot ulcers (IDFU) as well as to assess the association between PTX‐3 levels and IDFU severity. This study included 60 IDFU patients (Group 1), 45 diabetic patients without DFU (Group 2), and 45 healthy controls. Patients with IDFU were divided into mild, moderate, and severe subgroups based on classification of clinical severity. Patients who underwent amputation were also documented. Blood samples were collected to determine PTX‐3 levels. PTX‐3 levels in healthy controls, Group 1, and Group 2 were 5.83 (3.41‐20) ng/mL, 1.47 (0.61‐15.13) ng/mL, and 3.26 (0.67‐20) ng/mL, respectively. A negative correlation between plasma PTX‐3 and glucose levels was found. There were significant differences in terms of procalcitonin (PCT) and PTX‐3 levels in the subgroup analysis of Group 1. The PTX‐3 level in patients who did or did not undergo amputation was 4.1 (0.8‐13.7) and 1 (0.6‐15.1) ng/mL, respectively. Results suggest that PTX‐3 is a particularly effective marker in patients with IDFU, both in terms of predicting disease severity and assisting in the decision to perform amputation.

1. INTRODUCTION

Globally, diabetes mellitus (DM) is a widespread chronic metabolic disorder with an increasing prevalence rate among adult populations. Currently, DM is known to affect 415 million individuals, and this number is expected to rise to 642 million by 2040.1 Approximately 15% to 25% of all patients with diabetes develop foot ulcers at some time during their life span.2 These ulcers are responsible for the great majority of amputations in patients with diabetes.3, 4 Hyperglycaemia impairs the phagocytic and bactericidal function of neutrophils and leads to oxidative stress and an increase in inflammatory cytokines. Furthermore, hyperglycaemia induces the synthesis of tumour necrosis factor‐α, transforming growth factor‐β, interleukin‐6, and other inflammatory factors. These untoward consequences are associated with a predisposition for infection in diabetic foot ulcers.5, 6

Pentraxin‐3 (PTX‐3) and C‐reactive protein (CRP) are two structurally and functionally related molecules, both of which are acute‐phase reactants that contain five identical sub‐units known as the pentraxin family.7 In healthy individuals, PTX‐3 levels are closely related to gender, age, and blood sugar.8 PTX‐3, like CRP, is a fundamental component of the innate immunity and scavenging of necrotic/apoptotic cells.9, 10, 11 PTX‐3 is synthesised in mono‐nuclear phagocytic cells, myeloid dendritic cells, endothelial cells, fibroblasts, smooth muscle cells, and synovial cells.11, 12 Synthesis of PTX‐3 is induced by microbial components and inflammatory signals.10, 11 PTX‐3 locally produced at the site of inflammation has been associated with the severity of a variety of inflammatory disorders, as it can be detected very early during the course of the disease.11 A significant increase in the PTX‐3 level has been reported in viral and bacterial infections.13 Also, its role as a prognostic marker has been shown in community‐acquired pneumonia,14 ventilator‐associated pneumonia,15 febrile neutropenia,16 sepsis,13 inflammatory disorders, rheumatoid arthritis,17 tuberculosis, cardiovascular disorders, and kidney diseases.18, 19, 20 Conversely, PTX‐3 levels are negatively correlated with increasing insulin levels in response to increasing glucose concentrations.21

This study was undertaken to evaluate the diagnostic and prognostic values of PTX‐3 in patients with infected diabetic foot ulcers (IDFU) as well as to assess the association between PTX‐3 levels and IDFU severity. To the best of our knowledge, this study represents the first of its kind in patients with IDFU.

2. MATERIALS AND METHODS

The procedures in this single‐centre, prospective study were carried out in accordance with the principles of the Declaration of Helsinki, and the study protocol was approved by the Institutional Review Committee of Fırat University (date: 01 February 2018). Prior to collection of blood samples and clinical data, patients were provided information regarding the nature of the study, and written informed consent was obtained from each participant.

2.1. Study design

A total of 60 patients with IDFU (Group 1), 45 patients with type 2 DM without diabetic foot ulcers (non‐DFU, Group 2), and 45 age‐matched and sex‐matched healthy individuals as controls (Group 3) were included. The diagnosis of type 2 DM was based on the World Health Organisation criteria.22 Group 1 patients were classified by severity of IDFU as mild, moderate, and severe using the criteria of the Infectious Diseases Society of America International Working Group on the Diabetic Foot (IDSA‐IWGDF).23

Exclusion criteria included the suspicion of other infections, treatment with anti‐microbial or immunosuppressive agents or surgery within the past 6 months, malignancy, inflammatory disorders, and stage 1 disease based on IDSA‐IWGDF criteria.

2.2. Data collection

2.2.1. Clinical data

Age, gender, chronic diseases (cardiovascular disorders, peripheral vascular disease, diabetic complications etc.), amputation history, duration of hospital stay, and microbiologic and radiologic findings were retrieved from the hospital's electronic database. The decision for amputation was based on a clinical judgement of the infected body area and was performed only as a last resort.

2.2.2. Measurement of plasma PTX‐3 levels

Venous blood samples of 5 mL from each patient were placed in ethylenediaminetetraacetic acid tubes. Blood samples were centrifuged at 3500 rpm for 10 minutes within 40 minutes of sampling. Plasma samples were stored at −80° until the time of analysis. Serum PTX‐3 levels were analysed using an enzyme‐linked immunosorbent assay (ELISA) kit (Human PTX‐3; Catalogue no: 201‐12‐1939 Biological Technology Co., Ltd, Shanghai, China) according to the manufacturer's instructions.

The assay range for PTX‐3 ELISA kit was 0.08‐20 ng/mL, with an intra‐assay coefficient of variation of <12% and a minimum detection level of 0.051 ng/mL. Test results were expressed as ng/mL.

2.2.3. Routine biochemistry and haematology

Standard blood analyses including CRP, white blood cells (WBC) count, erythrocyte sedimentation rate (ESR), glucose, glycated haemoglobin (HbA1c), and procalcitonin (PCT) were performed in the Department of Clinical Biochemistry.

2.2.4. Statistical analyses

Data analysis was performed with SPSS 22 software (SPSS, Inc., Chicago, Illinois). The normal distribution of the variables was tested using visual methods (histograms, probability plots) and analytical methods (Kolmogorov‐Smirnov/Shapiro‐Wilk's test). For continuous variables with normal distribution, mean values (±SD) were provided, whereas those without normal distribution were expressed using median (min‐max). Categorical variables were expressed as frequency and percentage. For comparison of continuous variables with normal distribution, Mann‐Whitney U or Student's t test were used. For multiple comparisons, analyses were performed with one‐way ANOVA or Kruskal‐Wallis test. Categorical variables were compared with Pearson's χ ² and Fisher's exact test. For correlation analyses, Spearmen's rho correlation test was used. The cut‐off values of PTX‐3 and PCT for predicting the need for amputation were determined with a receiver operating characteristic (ROC) analysis. A P value of less than 0.05 was considered statistically significant.

3. RESULTS

In all, there were 79 (53%) female and 71 (47.3%) male participants in the study, with respective mean ages of 51.48 ± 17.24 and 55.28 ± 14.69 years (P > 0.05). PTX‐3 levels were 4.3 ng/mL in female patients and 3.7 ng/mL in male patients and no statistically significant difference was found between them (P = 0.78). However, there was no significant difference between the groups in terms of the mean age and gender (P > 0.05). Therefore, PTX‐3 levels between the groups are independent of the demographic data of our patients. Table 1 shows the distribution of the 150 subjects in study groups.

Table 1.

Distribution of demographic data in three study groups

	Group 1 (n: 60)	Group 2 (n: 45)	Group 3 (n: 45)	P value
Female/male	27/33	27/18	25/20	0.828
Agea	59.65 ± 10.89	57.65 ± 14.22	56.20 ± 11.18	0.530

Group 1: patients with infected diabetic foot ulcers; Group 2: diabetic patients with non‐diabetic foot ulcers; and Group 3: age‐matched and sex‐matched healthy subjects.

^aValues are expressed as mean ± SD.

PTX‐3 levels in healthy controls, Group 1, and Group 2 were 5.83 (3.41‐20) ng/mL, 1.47 (0.61‐15.13) ng/mL, and 3.26 (0.67‐20) ng/mL, respectively. Although healthy controls had significantly higher PTX‐3 compared with diabetic patients with IDFU and non‐DFU (P < 0.001 and P = 0.02, respectively), the latter two groups did not differ significantly (P = 0.18). Figure 1 shows the PTX‐3 levels and their association in study groups. Plasma PTX‐3, blood sugar, and HbA1c levels in Groups 1 and 2 are shown in Table 2.

Figure 1.

Plasma PTX‐3 levels and the associations in study groups^*. Controls, age‐matched, and sex‐matched healthy individuals; Non‐DFU, non‐diabetic foot ulcer, patients in Group 2; IDFU, infected diabetic foot ulcer, patients in Group 1

Table 2.

PTX‐3, blood glucose, and HbA1c levels in Group 1 and Group 2^a

	Group 1 (n: 60)	Group 2 (n: 45)	P value
PTX‐3	1.47 (0.61‐15.13)	3.26 (0.67‐20)	0.18
Gukoz	214 (75‐520)	158 (13‐407)	0.001
HbA1C	8.8 (5.2‐16.4)	7.6 (4.4‐13.6)	0.035

Abbreviations: HbA1c, glycated haemoglobin; PTX‐3, pentraxin‐3.

Group 1: patients with infected diabetic foot ulcers; Group 2: diabetic patients with non‐diabetic foot ulcers.

^aValues are expressed as median (min‐max).

A negative correlation between plasma PTX‐3 and glucose levels was found (P = 0.024, ρ = −0.231) (Figure 2), whereas no such correlation was found between HbA1c and PTX‐3 (P > 0.05).

Figure 2.

Association between plasma PTX‐3 and glucose levels (P = 0.024, ρ = −0.231). PTX‐3, pentraxin‐3

Table 3 shows the demographic and laboratory characteristics of the three subgroups of Group 1 patients. There were significant differences in terms of PCT and PTX‐3 levels in the subgroup analysis (P < 0.05). Additionally, there was a significant difference in CRP and ESR levels only between the mild versus moderate group and the mild versus severe group (P < 0.05), with no significant differences between moderate versus severe disease group.

Table 3.

Demographic and laboratory data in subgroups of Group 1

	Mild (n: 13)	Moderate (n: 26)	Severe (n: 21)	P value
Agea	59.4 ± 12.6	60.5 ± 8.9	61.5 ± 12.4	0.921
Female/male	6/7	11/15	10/11	0.952
CRPa	13.9 ± 12.2	79.1 ± 56.5	88 ± 68.3	0.001
WBC (1000/mm3)a	11.2 ± 4.8	12.9 ± 5.1	12.7 ± 3.2	0.559
ESRa	34.6. ± 19.5	64.4 ± 22	64.8 ± 29.5	0.003
Glukoza	252 ± 101	228 ± 87	225 ± 79	0.716
HbA1ca	8 ± 1.5	9.1 ± 2.4	9.7 ± 2.7	0.401
PCTb	0.12 (0.12‐0.15)	1.12 (0.12‐4.3)	2.14 (0.12‐3.5)	0.043
PTX‐3b	0.9 (0.6‐12)	2.1 (0.7‐15)	3.6 (0.8‐13.7)	0.011

Abbreviations: CRP, C‐reactive protein; ESR, erythrocyte sedimentation rate; HbA1c, glycated haemoglobin; PCT, procalcitonin; PTX‐3, pentraxin‐3; WBC, white blood cell.

^aThe values are presented as mean ± SD.

^bThe values are presented as median (min‐max).

Minor or major amputation was performed in 23 of the patients. 11 (47.8%) of these patients had transmetatarsal, 2 (8.6%) had tarsometatarsal, 9 (39.1%) had tibiotalar, and 1 (4.3%) had transfemoral amputation. The PTX‐3 level in patients who did or did not undergo amputation was 4.1 (0.8‐13.7) and 1 (0.6‐15.1) ng/mL, respectively, with statistical difference (P = 0.01). Also, no significant associations were found between CRP, ESR, and PCT (P > 0.05). Based on the ROC curve, the area under the curve (AUC) for predicting the need for amputation for PTX‐3 and PCT was 0.701 (P = 0.013) and 0.576 (P = 0.345), respectively (Figure 3, Table 4). According to the results of the likelihood ratio test, the cut‐off value for PTX‐3 was 3.63 ng/mL (sensitivity 57.1% and specificity 73.5%), and the cut‐off value for PCT was 0.26 ng/mL (sensitivity 33.3% and specificity 79.4%).

Figure 3.

Receiver operating characteristic curves of pentraxin (PTX‐3) and procalsitonin (PCT)

Table 4.

AUC and cut‐off values of PTX‐3 and PTC for predicting the need for amputation in patients with IDFU

	AUC	Cut‐off value	Sensitivity	Specificity	95% CI	P value
PTX‐3	0.701	3.63	0.571	0.735	0.562‐0.840	0.013
PCT	0.576	0.26	0.333	0.794	0.420‐0.732	0.345

Abbreviations: AUC, area under the curve; CI, confidence interval; IDFU, infected diabetic foot ulcer; PCT, procalcitonin; PTX‐3, pentraxin‐3.

Microbiological examination of the deep tissue samples obtained from patients showed the presence of Staphylococcus aureus in 18 patients (30%), Escherichia coli in 16 (26.6%), Klebsiella spp. in 10 (16.6%), coagulase negative Staphylococci in 8 (13.3%), Pseudomonas aeruginosa in 6 (10%), and Citrobacter freundii in two (3.3%).

4. DISCUSSION

Diabetic foot ulcers represent an important health problem with potentially serious consequences. Infections have been reported to occur in more than one‐half of diabetic foot ulcers.24 The presence of infection in the diabetic foot and propagation of the infection to the underlying bone are associated with increased rates of amputation.25, 26 In this study, the diagnostic and prognostic values of PTX‐3 as well as its association with IDFU severity were explored in patients diagnosed with IDFU.

A reduction in PTX‐3 was previously reported in patients with metabolic syndrome, which is associated with insulin resistance.27 In one study reported by Conles et al21 increasing insulin production was inversely correlated with PTX‐3 levels. Seman et al28 found reduced PTX‐3 in male patients with type 2 DM. In the current study, patients with DM, irrespective of the presence of IDFU, had lower PTX‐3 than healthy controls (P < 0.05). Also, a negative correlation between plasma glucose and PTX‐3 levels was found, supportive of the above‐mentioned results (P = 0.024, ρ = −0.231).

The diagnosis of IDFU should not be solely based on microbiological results,23 because the patient may rapidly deteriorate before the results of microbiological testing become available.29 Also, IDFU is associated with gangrene and lower extremity amputations.30 Thus, early and accurate diagnosis of IDFU is of utmost importance to reduce amputations and mortality.31 However, establishing a diagnosis of IDFU is not always straightforward,32 because the value of traditional laboratory markers such as ESR, WBC, or CRP has a limited value in the diagnosis of this condition.33 In contrast, procalcitonin (PTC) has a much higher diagnostic accuracy for bacterial infections.34 Because peak plasma concentrations of PTX‐3 are reached approximately 6 to 8 hours after an inflammatory insult in many pathological conditions, it also has diagnostic and prognostic roles.8, 11 In a study reported by Liu et al35 PTX‐3 was found to rise earlier than CRP in systemic inflammatory and infectious conditions and provided better diagnostic information than many biological markers, including CRP. Again, in the study by Park et al36 CRP and PCT levels were correlated with the severity of infection in DFU. Massara et al37 reported that PCT was an important biomarker in the diagnosis of patients with IDFU. The subgroup analyses in the current study showed significant differences in terms of PTX‐3 and PTC between mild, moderate, and severe disease subgroups (P < 0.05), in contrast with a significant difference in CRP and ESR only between mild versus moderate and mild versus severe groups (P < 0.05). In consideration of these data, the authors believe that PTX‐3, like PCT, may represent an early marker for predicting the clinical course of patients with IDFU.

DM is associated with a vast number of complications, including foot ulcerations that may require lower extremity amputation. Around 20% of patients with IDFU undergo lower extremity amputation.38 In a study reported by Reiner et al39 patients with elevated PCT levels at baseline were more likely to require subsequent amputations. Conversely, although Karakas et al40 could not find significant PCT and CRP differences between patients who did or did not have amputation, a significant elevation in ESR was found (P = 0.042). In the current study, 39% of the patients underwent amputation. Patients undergoing amputation had significantly higher elevation of PTX‐3 than patients who did not have amputation (P = 0.01), whereas other inflammatory parameters were not significantly different. In this study, the cut‐off value of PTX‐3 for predicting the need for amputation was 3.63 ng/mL. On the basis of these observations, the authors believe that increasing PTX‐3 levels may influence the decision of surgeons to perform amputation or to salvage the extremity.

This study is the first to suggest a diagnostic and prognostic role for plasma PTX‐3 levels in patients with IDFU. However, there are certain limitations of this study, including the small sample size, the absence of a group consisting of patients with uninfected diabetic foot ulcers, and assessment of the severity of IDFU using only the IDSA‐IWGDF clinical classification, which may have resulted in an inter‐observer variability in the severity assessments.

Patients with DM have a predisposition for increased risk of infections. As such, prognostic markers are required in patients with IDFU. In conclusion, results suggest that PTX‐3 levels are lower in patients with IDFU compared with healthy controls and may represent a valuable marker both for predicting disease severity and guiding the decision to amputate. Further studies are warranted to confirm these observations.

CONFLICT OF INTEREST

The authors have no conflicts of interest to report.

Ozer Balin S, Sagmak Tartar A, Uğur K, et al. Pentraxin‐3: A new parameter in predicting the severity of diabetic foot infection? Int Wound J. 2019;16:659–664. 10.1111/iwj.13075