• Open Access

Synovial Fluid D-Dimer Concentration in Foals with Septic Joint Disease

Authors

  • T. Ribera,

    1. Servei de Cirurgia Equina, Universitat Autònoma de Barcelona, Barcelona, Spain; and the Laboratory of Biostatistics & Epidemiology, Facultat de Medicina, Universitat Autònoma de Barcelona, Statistics & Methodology Support Unit, IDIBAPS, Hospital Clínic, Barcelona, Spain
    Search for more papers by this author
  • L. Monreal,

    1. Servei de Medicina Interna Equina, Universitat Autònoma de Barcelona, Barcelona, Spain; and the Laboratory of Biostatistics & Epidemiology, Facultat de Medicina, Universitat Autònoma de Barcelona, Statistics & Methodology Support Unit, IDIBAPS, Hospital Clínic, Barcelona, Spain
    Search for more papers by this author
  • L. Armengou,

    1. Servei de Medicina Interna Equina, Universitat Autònoma de Barcelona, Barcelona, Spain; and the Laboratory of Biostatistics & Epidemiology, Facultat de Medicina, Universitat Autònoma de Barcelona, Statistics & Methodology Support Unit, IDIBAPS, Hospital Clínic, Barcelona, Spain
    Search for more papers by this author
  • J. Ríos,

    1. Departament de Medicina i Cirurgia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, Barcelona, Spain; and the Laboratory of Biostatistics & Epidemiology, Facultat de Medicina, Universitat Autònoma de Barcelona, Statistics & Methodology Support Unit, IDIBAPS, Hospital Clínic, Barcelona, Spain
    Search for more papers by this author
  • M. Prades

    1. Servei de Cirurgia Equina, Universitat Autònoma de Barcelona, Barcelona, Spain; and the Laboratory of Biostatistics & Epidemiology, Facultat de Medicina, Universitat Autònoma de Barcelona, Statistics & Methodology Support Unit, IDIBAPS, Hospital Clínic, Barcelona, Spain
    Search for more papers by this author

  • Preliminary results from this study were presented at the 9th Congress of the European Veterinary Emergency and Critical Care Society (EVECCS), in Cambridge, UK, June 2010. The manuscript represents a portion of a thesis submitted by the first author at the Universitat Autònoma de Barcelona as partial fulfilment of the requirements for her PhD program, supervised by the last authors.

Corresponding author: Thaïs Ribera, DVM, Servei de Cirurgia Equina, Departament de Medicina i Cirurgia Animals, Facultat de Veterinària, Universitat Autònoma de Barcelona, 08193-Bellaterra, Barcelona, Spain; e-mail: thais.ribera@uab.cat

Abstract

Background: Increased synovial fibrinolytic activity (detected by increases in synovial D-Dimer concentrations) has been observed in different joint diseases in humans and adult horses, presumably in order to minimize fibrin deposition within the joint and thus avoid its detrimental effects.

Objective: To investigate fibrinolytic pathway activation in joint sepsis in foals by measuring synovial D-Dimer concentrations.

Animals: Eighteen septic foals with septic joints, 9 septic foals without septic joints, 9 systemically healthy foals with septic joint, and 3 controls are included.

Methods: Prospective observational clinical study of foals admitted for septic arthritis. Synovial D-Dimer concentration and routine synovial fluid analysis were performed. Diagnosis of joint sepsis was made whenever synovial total nucleated cell count was >30,000 cells/μL, synovial total protein >4 g/dL, and neutrophil percentage of >80%, or synovial fluid culture resulted positive. Results were compared among groups by general lineal models.

Results: Synovial D-Dimer concentration was significantly (P < .001) higher in the foals with septic joints compared with foals without joint disease (P < .001).

Conclusions and Clinical Importance: Septic joint disease is associated with a marked increase of synovial D-Dimer concentration (marked activation of the fibrinolytic activity) within the affected joint. Although further studies are needed, the measurement of synovial D-Dimer concentration may be considered a complementary diagnostic marker of septic joint disease.

Abbreviations:
CF

control foals

FSJ

systemically healthy foals with a septic joint

OA

osteoarthritis

OCD

osteochondritis dissecans

SF

septicemic foals without septic joints

SFSJ

septic foals with septic joints

TNCC

total nucleated cell count

TP

total protein concentration

Septic arthritis and especially septic polyarthritis in septicemic foals1,2 are disease entities that can result in degenerative joint disease3 and decrease the likelihood of athletic performance.4 A retrospective study involving Thoroughbred foals showed that the diagnosis of septic arthritis reduced racing starts.4 Early detection and treatment of joint sepsis might improve the prognosis in these sick foals. Synovial fluid total nucleated cell count (TNCC), total protein concentration (TP), and cell differential are the most useful parameters in making the diagnosis of septic arthritis.5 Occasionally, it might be difficult to differentiate an early stage of infective arthritis with a relatively low TNCC from a traumatic arthritis with a high white cell count6 or from animals that have been treated with antibiotics and may have a negative result on culture. In these cases having a synovial biomarker test that is affordable and quick could be useful in the decisions for therapy. Use of biomarkers focused on diagnosing infectious joint diseases has been reported.7–10

Concurrent activation of synovial fibrinolytic and coagulation pathways has been detected as a result of joint inflammation in human arthritic patients and especially so in rheumatoid arthritis patients.11 Activation of the fibrinolytic pathway is considered a protective, homeostatic response in inflamed joints. In degenerative joint diseases such as osteoarthritis (OA), the activation of both pathways is also apparent in humans11 and it has been recently observed in horses affected by osteochondritis dissecans (OCD) and OA.12

D-Dimer concentration measurement is a sensitive marker for systemic and local (ie, peritoneal fluid) fibrinolytic activity.13,14 Hence, any increase of the synovial fluid D-Dimer concentration is considered strongly related to the activation of the synovial fibrinolytic activity in affected joints.15 The main purpose of this study was to assess synovial fibrinolytic activity of foals with septic arthritis by the determination of synovial D-Dimer concentrations. Our hypotheses were that (1) the synovial D-Dimer concentration is markedly and significantly higher in foals with septic arthritis than in those foals without joint disease and (2) this increase of synovial D-Dimer concentration in septic joints is not affected by the presence of systemic disease.

Materials and Methods

In this prospective observational clinical study, foals admitted to the Unitat Equina-Fundació Hospital Clínic Veterinari between October 2006 and September of 2009 were distributed into 4 groups as follows: (1) septic foals with septic joints (SFSJ), (2) septic foals without septic joints (SF), (3) systemically healthy foals with a septic joint (FSJ), and (4) control foals (CF).

The diagnosis of septic joint disease was reached when any one of the following findings was present: (a) the synovial TNCC was >30,000 cells/μL, synovial TP > 4 g/dL, and neutrophil percentage of >80% in the synovial cytology,16 (b) presence of microorganisms in the synovial cytological examination, or (c) a positive synovial fluid culture. Diagnosis of septicemia was confirmed when any of the following findings were present: a positive blood culture, sepsis score ≥11,17 septic polyarthritis, and/or postmortem findings consistent with a septicemic foal. The diagnosis of septic polyarthritis was confirmed when at least 2 joints were determined to be septic.

Three healthy foals that were admitted to the hospital for orthopedic problems unrelated to joint disease and that were subsequently euthanized because of financial restraints or a poor athletic prognosis were also included in the study as controls.

Synovial Fluid Collection and Analysis

Synovial fluid samples were aseptically collected from joints either (1) during initial foal examination before any form of treatment or (2) immediately after euthanasia. Samples were placed in 1-mL tubes containing EDTA for a routine synovial fluid analysis (color, clarity, mucin clot formation, TNCC, TP, and cytology evaluation) under the supervision of a boarded clinical pathologist. An automated blood cell countera was used for TNCC determination in synovial fluid, and TP was measured by refractometry. Cytospinb preparations and Diff-Quick staining were used for the microscopic evaluationc of synovial fluid cytologies. Samples collected after hours were immediately analyzed on site by a semiautomatic cell counter,d and kept refrigerated at 4°C for later submission to the laboratory for cytologic examination.

Synovial D-Dimer Concentration (ng/mL)

Synovial fluid samples (1 mL) were placed in 1 mL tubes containing sodium citrate (3.8%) (1 : 9, citrate: sinovial fluid). Within the first 2 hours after collection, samples were centrifuged at 1,000  × g for 10 minutes and supernatants were pipetted and stored at −70°C for further analysis. Synovial D-Dimer concentration was measured in duplicate, by a quantitative immuno-turbidimetric latex agglutination assaye with commercial reagents and controls,f according to the manufacturer's instructions. The upper and lower limits of detection of the assay are 5,000 and 0 ng/mL, respectively. Samples higher than 5,000 ng/mL were diluted 1 : 4 with saline solutionf as many times as necessary. Results obtained were multiplied by the dilution factor. This assay has been previously validated and employed by our group in different samples of equine origin.13,14,18–20a,g

Statistical Analysis

Quantitative results for principal variables (synovial D-Dimer concentration, TNCC, and TP) and age were expressed as a median and interquartile range (25th–75th percentile). For the statistical assessment of sepsis, as data did not follow a normal distribution, a nonparametric approach, by means of rank-transformation, was used with a general lineal model (GLM) with a posthoc analysis for evaluation of differences between disease groups. In the case of qualitative variables (sex or survival foals), the description was based on absolute frequencies and percentages.

A two-tailed type-I error of 0.05 was used in all statistical inference. Results were considered to be significant when P < .05. Commercial software (SPSS for Windows, version 15.0)h was used for the statistical analysis.

Results

Animals Included in the Study

Thirty-nine foals were included in the study: 18 septic foals with septic joints, 9 septic foals without septic joints, 9 FSJ, and 3 controls.

Sex, Breed, and Age. Twenty of 39 foals were males and 19 were females. Breed distribution was as follows: Andalusians (24), Arabians (6), crossbreds (3), ponies (2), Westfalian (1), Quarterhorse (1), standardbred (1), and warmblood (1). The distribution of ages by diagnosis was as follows: 6 days (3–14 days) for the SFSJ; 0.5 days (0.5–4 days) for the SF, 25 days (22–34 days) for the FSJ, and 76.5 days (15–138.7 days) for the CF.

SFSJ

All of these 18 septic foals with septic polyarthritis had heat and synovial fluid effusion in at least 2 joints (range 2–7 joints). Six of these foals had mild to severe lameness, and the remaining 12 foals were nonambulatory or too weak to be able to appreciate lameness.

Other diagnoses, together with the septic polyarthritis related to septicemia, in this group of foals included pneumonia (6), enteritis (5), omphalitis with or without patent urachus (5), and failure of transfer of passive immunity (5).

Seven blood cultures were positive: Escherichia coli (3), Klebsiella spp. (2), Pasteurella spp. (1), Pseudomonas putida (1); 5 were negative and in the remaining 6 foals blood culture was not performed as they had received antibiotics before admission.

Five synovial fluid cultures resulted positive: E. coli (1), Klebsiella pneumonia (1), Bacteroides fragiles (1), Salmonella spp. (1), Pasteurella spp. (1), and 3 resulted negative.

Septic Foals without Septic Joints (SF)

None of the SF presented clinical evidence of joint disease. Results of the standard synovial fluid analysis from the samples collected (synovial TNCC and TP) were all within normal ranges (<500–1,000 nucleated cells/L and <2.5 g/dL, respectively). Synovial samples from these foals were collected immediately after euthanasia, except for 2 foals where joint involvement was initially suspected but was later ruled out. The sepsis score was ≥11 in all of them. Major diagnoses confirmed other than septicemia were failure of transfer of passive immunity (4), immaturity (3), and hypoxic-ischemic encephalopathy (2).

FSJ

All 9 foals showed marked to severe lameness (range 3–5/5) and synovial effusion in all affected joints. The cause of the septic arthritis was confirmed to be traumatic in only 1 foal attributable to a puncture wound. Their sepsis scores on admission were always <10. Clinical examinations were normal, except for the presence of lameness and distension of affected joints.

None of the 6 blood cultures performed resulted positive.

Three out of 6 synovial cultures resulted positive (E. coli [2] and Klebsiella spp. [1]).

CF

The 3 foals that were included as controls were referred for open fetlock luxation, tibial fracture, and pastern fracture, respectively. Their clinical examinations were normal, except for the main orthopedic problem, and their postmortem examination confirmed no joint abnormalities or presence of other systemic disease.

Excluded Foals. Eleven other foals were excluded from the study because their final diagnoses were consistent with nonseptic inflammatory joint disease (such as OCD, chip fracture, and Rhodococcus equi immunomediated polyarthritis) and did not meet the inclusion criteria of our study.

Outcome. Fifteen out of 36 foals (42%) survived and the remaining 21 foals (58%) did not. Seven of the 15 survivors were FSJ (47%), 6 were SFSJ (40%), and 2 were SF (13%). Twelve of the 21 nonsurviving foals were SFSJ (57%), 7 were SF (33%), and 2 were FSJ (9%).

Synovial Fluid Samples Included in the Study. A total of 65 synovial fluid samples were analyzed: 42 septic synovial fluid samples from the SFSJ, 10 nonseptic synovial fluid samples from the SF, 9 septic synovial fluid samples from the FSJ, and 4 normal synovial fluid samples from the CF.

Foals included in the study grouped by diagnosis, number of synovial fluid samples collected, bilateral or unilateral collection, and specific joint sampled are shown in Table 1.

Table 1.  Results of the number of foals, number of synovial fluid samples collected, bilateral or unilateral collection, and specific joint sampled of the foals in the study: septic foals with septic joints (SFSJ), septic foals without septic joints (SF), systemically healthy foals with a septic joint (SJF), and control foals (CF).
DiagnosisNumber of FoalsNumber of Synovial Fluid SamplesSample Collection (Number of Foals)Sampled Joints
  1. FP = femoropatellar joint; FT = femorotibial joint; MC = middle carpal joint; MCP = metacarpophalangeal joint; TC=tarsocrural joint; RC = radiocarpal joint.

SFSJ1842Unilateral (24)
Bilateral (9)
11 TC, 3 MC, 2 RC, 4 MCP, 1 FT, 2 FP, 1 elbow
10 TC, 6 MC, 2 RC
SF910Unilateral (8)
Bilateral (1)
6 RC, 2 TC
1 TC
FSJ99Unilateral (9)4 TC, 2 FP, 1 MC, 1 RC, 1 FT
CF34Unilateral (2)
Bilateral (1)
2 RC
1 MC

Standard Synovial Fluid Analysis

Synovial TNCC and TP were both significantly higher in the SFSJ and FSJ compared with the SF and with the CF (P < .001). Total proteins were also significantly higher in the SF compared with the CF (P = .006) (Table 2).

Table 2.  Data regarding the principal variables of interest: synovial total nucleated cell count (TNCC), total protein (TP) and D-dimer concentration, grouped by septic foals with septic joints (SFSJ), septic foals without septic joints (SF), systemically healthy foals with a septic joint (FSJ), and control foals (CF).
 Synovial Fluid TNCC (cells/μL)Synovial Fluid TP (g/dL)Synovial Fluid D-Dimer Concentration (ng/mL)
  • Values given are expressed in (median, interquartile range). Analyses were performed by means GLM models with non-parametric approach.

  • *

    Significantly different from SF and CF (P < .001).

  • $

    Significantly different from SFSJ and FSJ (P < .001) and from SF (P = .006).

  • &

    Significantly different from SFSJ and FSJ (P < .001) and from CF (P = .006).

SFSJ13,600 (4,450–60,400)*3.5 (2.5–4.1)299,104 (122,112–687,232)*
SF500 (450–850)1.80 (0.9–2.2)&40,640 (19,232–95,512)
FSJ14,025 (8,500–50,200)*3.4 (3.2–4.0)400,384 (256,768–632,064)*
CF600 (500–800)0.85 (0.8–1.4)$36,753 (30,096–45,041)

Synovial D-Dimer Concentration (ng/mL). The synovial D-Dimer concentration in SFSJ and FSJ was significantly (P < .001) higher when compared with the SF and to the CF (Fig 1).

Figure 1.

 Box-plots of synovial D-Dimer concentrations from the controls (CF), septic foals without septic joints (SF), systemically healthy foals with a septic joint (FSJ), and septic foals with septic joints (SFSJ), respectively. Central boxes represent the values from the lower to upper quartile (25–75 percentile), while the middle line represents the median. Significant differences are shown by their specific P values.

The GLM models showed that septicemia did not influence synovial D-Dimer concentration of the affected foals.

Control and nonseptic synovial samples required dilutions between 1 : 4 and 1 : 16. Septic synovial samples, however, required at least 1 : 64 or 1 : 256 dilutions. Exceptionally, a few dilutions 1 : 1,024 were required in severely infected joints.

Discussion

The results of this study confirm that septic joint disease in foals is associated with a strong activation of the synovial fibrinolytic pathway based on the markedly and significantly higher synovial D-Dimer concentrations found in synovial fluid of septic joints when compared with those measured in noninfected joints of either septic or healthy foals.

This activation of the fibrinolytic pathway seen in synovial fluid from foals with septic arthritis is much more significant than what has been previously reported in adult horses with different nonseptic arthritisa and similar or even higher than the synovial D-Dimer values of human inflammatory arthritides.11 A study with human arthritic patients showed that plasmatic and synovial D-Dimers were significantly higher in inflammatory arthritides than in OA (76,287 ± 16,360 ng/mL, synovial data) and that the synovial levels were highest in rheumatoid arthritis patients (407,006 ± 46,610 ng/mL).11 The study concluded that activation of the coagulation and fibrinolytic cascades was evident in both inflammatory and degenerative joint diseases. In the equine studya results showed that synovial D-Dimer concentration of horses with OCD (25,172 (8,472–73,472 ng/mL) and OA (21,440 (6,641–50,092 ng/mL) were both higher compared with controls (12,858 (5,794–17,780 ng/mL) but were only significantly different in horses with OCD. Plasma D-Dimer values were not evaluated. Because the highest synovial D-Dimer concentrations in both human11 and equine patients (foals and adultsa) have been observed in the arthritides with the strongest inflammatory component: rheumatoid arthritis, septic arthritis, and OCD, respectively, the most likely cause for such increased activation of the coagulation and fibrinolytic pathways is related to the severity of the inflammation.

Baseline D-Dimer concentrations of normal foals could be slightly more increased than in adult horses but a larger population is needed to confirm this. Additionally, as the range of age in our foal population was small enough between the different groups, a statistical analysis adjusted by the foals' age was considered unnecessary.

Plasma overspill of D-Dimer concentration into the joints of very sick foals was ruled out because synovial D-Dimer concentration in septic foals without septic joint disease was much lower than the concentration observed in SFSJ and therefore increased synovial D-Dimer concentration should only be attributed to local joint inflammation and not to systemic disease. Furthermore, the synovial D-Dimer concentration in septic joints was much higher (approximately from 100 to 150-fold higher) than concentrations previously reported from the plasma of very sick foals,18,19 or even from samples with blood contamination.14

The fact that septicemia was not diagnosed in the FSJ does not rule out bacterial hematogenous seeding as the initial cause of the single septic arthritis in these foals, especially when considering that joint infection is the most common consequence to bacteremia in newborn foals and that penetrating articular wounds rarely occur in these animals (<60 days).1

Regarding standard synovial fluid variables (TNCC and TP), these were significantly higher compared with the nonseptic joint groups (SF and CF), as expected. Septicemic foals without joint disease, however, also showed significantly higher total proteins compared with the controls, but these differences were not relevant as the total protein values in both groups were within normal ranges (<2.5 g/dL).

A limiting factor in this study is that the group of CFs is small. However, considering that plasma overspill of D-Dimers from systemically sick patients leads to imperceptible changes in the synovial D-Dimer concentration, septicemic neonates without septic joint diseases can be considered similar to controls, although we are aware that a larger population of controls would be highly recommended.

Because the foal population in the present study is small, it does not allow the establishment of a synovial D-Dimer cut-off concentration to diagnose joint infection, although it is evident that an extremely increased synovial D-Dimer concentration is highly suggestive of septic joint disease. Further clinical trials need to be performed to be able to determine this cut-off value.

In conclusion, synovial D-Dimer concentration is extremely high in foals with septic joints, which shows that the inflammatory response attributable to infection produces a strong activation of fibrinolytic activity in the affected joints. Although further studies are warranted, the measurement of synovial D-Dimer concentration may prove in the future to be an additional marker of septic joint disease. The test is quick, easy, patient-side, and affordable in clinical practice.

Footnotes

a Advia 120 Analyzer, Bayer Lab, New York, NY

b Shandon CytoSpin III Cytocentrifuge, GMI Inc, Ramsey, MN

c Microscopy Hemacolor, Merck Egaa, Darmstadt, Germany

d Vet ABC Diff Hematology Analyzer, Heska, Fort Collins, CO

e Miniquant, Biopool, Trinity Biotech, Wicklow, Ireland

f Miniquant-1, Biopool, Trinity Biotech

g Cesarini C, Monreal L, Segura D, et al. Hemostatic follow up of horses with medical and surgical colic. J Vet Intern Med 2009;23:434 (abstract)

h SPSS Inc, Chicago, IL

Ancillary