MURDOCH RESEARCH REPOSITORY

This is the author’s final version of the work, as accepted for publication following peer review but without the publisher’s layout or pagination.

The definitive version is available at http://dx.doi.org/10.1016/j.jevs.2015.10.004

Smuts, C., Mills, J., Myles, R. and Gaál, T. (2016) Lactate dehydrogenase activity in abdominal fluid from horses with colic. Journal of Equine Veterinary Science, 36 . pp. 58-62.

http://researchrepository.murdoch.edu.au/29488/

Copyright: © 2015 Elsevier Inc.

Accepted Manuscript

Lactate dehydrogenase activity in abdominal fluid from horses with colic

Celia Smuts, Jennifer Mills, René Myles, Tibor Gaàl

PII: S0737-0806(15)00612-7

DOI: 10.1016/j.jevs.2015.10.004

Reference: YJEVS 1966

To appear in: Journal of Equine Veterinary Science

Received Date: 24 July 2015

Revised Date: 13 October 2015

Accepted Date: 15 October 2015

Please cite this article as: Smuts C, Mills J, Myles R, Gaàl T, Lactate dehydrogenase activity inabdominal fluid from horses with colic, Journal of Equine Veterinary Science (2015), doi: 10.1016/j.jevs.2015.10.004.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service toour customers we are providing this early version of the manuscript. The manuscript will undergocopyediting, typesetting, and review of the resulting proof before it is published in its final form. Pleasenote that during the production process errors may be discovered which could affect the content, and alllegal disclaimers that apply to the journal pertain.

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

Lactate dehydrogenase activity in abdominal fluid from horses 1

with colic 2

Celia Smutsa*, Jennifer Millsa, René Mylesa, Tibor Gaàla 3

aDepartment of Pathology, Veterinary and Life Sciences, Murdoch University, 4

South Street, Murdoch, 6150, Western Australia 5

*Corresponding author: celiasmuts@gmail.com 6

7

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

Abstract: 8

The purpose of this study was to determine whether lactate dehydrogenase 9

(LDH) activity in abdominal fluid could be used as a prognostic indicator in 10

horses with colic. 11

LDH activity was measured in 27 abdominal effusions from horses with colic 12

presented to Murdoch University veterinary teaching hospital using three 13

different LDH test methods. Lactate in effusions was also measured in 11 of the 14

horses. 15

LDH activity was significantly different for each test methods used - the ratio of 16

Randox wet chemistry LDH lactate to pyruvate (L-P): Randox wet LDH P-L:IDEXX 17

dry chemistry P-L was approximately 1:2:4. LDH activity in the abdominal 18

effusions was significantly higher with all methods in the horses that died or 19

were euthanased due to abdominal sepsis or advanced neoplasia than in those 20

that survived following treatment for colic signs due to mechanical obstructions 21

or non-septic abdominal inflammation. LDH activity showed moderate to good 22

correlation (r=0.73-0.86) with lactate concentration of the fluid. 23

In conclusion, LDH activity in abdominal fluid may be a useful prognostic test in 24

horses with colic. The test method for LDH measurement must be known and 25

remain constant for meaningful interpretation. Significantly higher levels of LDH 26

activity may be present in horses with colic due to sepsis or advanced neoplasia 27

than in those with colic due to non-septic inflammation or mechanical 28

obstructions, that may respond to treatment. 29

Keywords 30

Equine colic, abdominal lactate dehydrogenase activity, LDH, prognosis 31

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

1. Introduction 1

Although LDH activity measurement is used extensively in human medicine to 2

differentiate mainly between transudative and exudative pleural effusions [1-4], 3

few studies have been carried out on the body cavity effusions of veterinary 4

patients [5-7]. 5

Lactate dehydrogenase (LDH) is an enzyme that is present in the cytoplasm of 6

almost all cells, including leukocytes [8] and red blood cells, and is an end 7

enzyme in the glycolysis pathway. It acts as a hydrogen transfer enzyme, 8

catalysing the oxidation of L-lactate to pyruvate, which is the final step in the 9

metabolic chain of anaerobic glycolysis. The reaction is reversible and the 10

reaction equilibrium favours the reduction of pyruvate to lactate [9, 10]. LDH 11

can be measured using both the forward and reverse reactions, using both wet 12

chemistry (conventional) and dry (solid phase) chemistry methods. 13

On average human tissues have about 500 times the total LDH levels found in the 14

serum, with very high levels in the liver (9000 IU/g), heart (25000 IU/g), skeletal 15

muscle (9000 IU/g) and lung (9500 IU/g) [11]. Enzymes such as LDH appear to 16

serve no function in body cavity effusions, but can serve as indicators of 17

disturbed cellular integrity induced by pathological conditions [11]. Even a 18

small amount of tissue damage can result in significant elevation of LDH activity 19

and its extracellular appearance can therefore be used to detect cell injury or 20

death [10]. Increased activities of LDH in peritoneal fluid may indicate 21

inflammation and tissue damage including ischaemia [12]. Lactate can also be 22

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

measured in abdominal fluid on lab-based and point-of care monitors and is 23

frequently used for evaluating abdominal fluid in horses with colic [13]. 24

This study explores the use of lactate dehydrogenase activity of abdominal 25

effusions as an aid in determining prognosis in horses with colic. 26

2. Materials and methods 27

Abdominal fluid was collected from 27 horses presented to the Murdoch 28

University Veterinary Hospital and submitted to the MUVH clinical pathology 29

laboratory for evaluation. Samples submitted to the laboratory were often from 30

referral cases and in cases wherein the clinicians required more information 31

than their in-house analysis provided, e.g. prolonged colic, complicated medical 32

or surgical cases or insurance cases. Samples were excluded if they were 33

analysed later than 24 hours after collection. Use of samples complied with the 34

Murdoch University animal ethics protocols for excess samples obtained for 35

diagnostic purposes. 36

Lactate was also measured in the abdominal samples from 11 of the horses on a 37

radiometer ABL 700 blood gas analyzer (Radiometer Medical ApS, Denmark) 38

using samples collected into EDTA anticoagulant and analysed within 15-30 39

minutes after collection. 40

Routine analysis was performed on all submitted abdominal effusions. This 41

included obtaining a total nucleated cell count and red cell count by flow 42

cytometry (Advia® 120 haematology analyser, Siemens Health Care Diagnostics, 43

USA) on the effusion sample. Packed red cell volume (PCV) was measured in 44

haematocrit capillary tubes following centrifugation. Total protein was 45

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

3

measured on the supernatant of the effusion using a temperature compensated 46

refractometer (Reichert Vet 360, NY, USA). 47

A direct smear and a cytocentrifuged preparation (50 and 200μL of abdominal 48

fluid, depending on visual density of the sample) were air dried and stained with 49

Wright-Giemsa (HEMA-TEK modified Wright-Giemsa stain pack, Bayer, 50

Germany) and evaluated microscopically by a clinical pathologist. Culture and 51

sensitivity were performed by an independent laboratory (Vetpath, Perth, 52

Western Australia) on samples with a high neutrophil percentage (>70%), 53

presence of micro-organisms observed visually on microscopy and/or high 54

clinical suspicion of infection. 55

An aliquot of each abdominal effusion sample was spun at 3000rpm (1500 G) in 56

a Jouan CR3i multifunction centrifuge (Thermo Fisher Scientific, USA) at 4˚C for 57

10 minutes, the supernatant placed in a 1ml plastic tube and the sediment 58

discarded. The supernatant was kept at 4˚C and analysed for lactate 59

dehydrogenase (LDH) activity within 24 hours of sample submission. LDH was 60

measured on each sample using three different methods. Firstly LDH was 61

measured via two wet chemistry (traditional) methods using the pyruvate to 62

lactate (LDH P-L) and lactate to pyruvate reactions (LDH L-P) using a Randox 63

Daytona biochemistry analyser (Northern Ireland, UK) and two commercial 64

Randox reagents (catalogue numbers 3818 and 3842, respectively). If results 65

obtained from samples were out of analytic range, the samples were diluted 1 in 66

20 in distilled water (50μL of sample in 950μL of distilled water) and the 67

analysis repeated. 68

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

4

LDH P-L activity was also measured using dry-slide technology on a VetTest® 69

Chemistry analyser (IDEXX laboratories, USA). This test was run on undiluted 70

samples if the wet chemistry results did not require dilution, or the diluted 71

sample if dilution had been required for the Randox Daytona analyser. 72

Bland-Altman plots, regression analysis and one-way analysis of variance 73

(ANOVA) functions in Excel 2011 for Mac (Microsoft, USA) and StatPlus:mac LE 74

(v5.9.20, AnalystSoft Inc) were used to determine agreement and statistical 75

differences between different methods of LDH measurement. The final clinical 76

diagnosis and outcome were recorded for each horse, noting whether it survived 77

(>1 month) or succumbed to the condition causing colic. Independent sample t-78

tests were used to compare LDH activity between horses that died due to sepsis 79

or neoplasia and those that survived following treatment with or without 80

surgery (mostly with non-septic inflammation or mechanical intestinal 81

obstruction). Receiver operating characteristic (ROC) curves for determining 82

cut-off values between survival and death were produced using Medcalc 83

(Version 12.3.0.0, MedCalc Software, Mariakerke, Belgium). 84

3. Results 85

LDH activity was measured in the abdominal fluid of 27 horses with colic, 19 of 86

which survived and 8 that died or were euthanased due to abdominal sepsis or 87

neoplasia (Table 1). 88

LDH activity was significantly higher in the latter group (Randox wet LDH L-P: 89

p=0.039; Randox wet LDH P-L: p=0.040; IDEXX dry LDH P-L: p=0.039) (figure 1). 90

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

5

The different methods of testing LDH activity provided significantly different 91

results from each other (ANOVA, p=0.028), however, there was good agreement 92

between the different methods (figure 1). The ratio of Randox wet LDH L-93

P:Randox wet LDH P-L:IDEXX dry LDH P-L was almost 1:2:4 from all samples. 94

LDH was significantly increased using all three methods of LDH measurement in 95

horses that died compared to those that survived (p<0.05, figure 2). There was 96

no significant difference in LDH activity between colic horses that survived 97

without surgery (n=11) compared to the 8 horses that survived but where 98

surgery was considered necessary (p=0.37-0.52 for all three methods of 99

measuring LDH activity) or between horses that died due to neoplasia (n=3) or 100

sepsis (n=5, p>0.71). 101

A cut-off value for all methods was attempted using ROC curves to aid in 102

determining prognosis for horses with colic in our study population based on 103

LDH activity (Table 2). 104

Lactate concentration was also measured in abdominal effusions from 11 horses, 105

10 of which recovered and 1 that died with a septic peritonitis. There was 106

moderate correlation of lactate concentration with wet LDH L-P and wet LDH P-L 107

activity (r = 0.73) and good correlation with IDEXX dry chemistry LDH P-L 108

(r=0.86). 109

4. Discussion 110

In this small study, we found that LDH activity in abdominal effusions may 111

provide information regarding the prognosis in horses with colic, and may be 112

useful for identifying sepsis or neoplasia, although further comparison with 113

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

6

abdominal fluid lactic acid levels is required to check for its added value both 114

from a practical and scientific point of view. Very high LDH activity was usually 115

associated with sepsis (due to intestinal leakage or rupture) or advanced intra-116

abdominal neoplasia, with all but one septic/neoplastic effusion (1 of 9) 117

resulting in death or euthanasia, and the higher the activity the more likely it was 118

that sepsis or neoplasia was present. 119

Three methods of LDH measurement were compared in this study, and they 120

resulted in significantly different values, although they did follow a similar trend. 121

The cut-off values mentioned in this study depend on the test method used. It is 122

therefore essential to know which methods for measuring LDH activity was used 123

and to use one method consistently. For practitioners an in-house dry chemistry 124

test may be more relevant (the IDEXX dry chemistry is an in-house method of 125

rapid analysis), though larger laboratories usually use the wet chemistry 126

(traditional) analysis. LDH activity can be measured on samples immediately 127

after centrifugation to remove particulate matter. Only two analysers and three 128

test kits were used in this study, and further investigation should be done into 129

other available test kits and analysers. 130

We found moderate to good correlation between LDH activity and lactate 131

concentration in abdominal fluid from horses with colic. A disadvantage of 132

lactate is the lability of the molecule, therefore its concentration must be 133

measured soon after sample collection for accurate results (point-of-care devices 134

are available) [14]. We found LDH to be stable for up to 3-7 days in a 135

refrigerator, as reported by others [9, 10], which allows a little more flexibility 136

between sample collection and testing. Measuring LDH activity may therefore be 137

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

7

a useful adjunct to routine laboratory and in-house analysis of effusion fluid, 138

clinical signs and effusion lactate concentration [15, 16], especially where an on-139

site lactate meter is not available, or possibly to evaluate retrospectively how a 140

condition has progressed, though more investigation of ongoing disease is 141

required for this. 142

The reason for the high proportion of horses with gastro-intestinal rupture and 143

neoplasia in this group may be that they were drawn largely from a referral 144

population. One of the shortcomings of this study is that there are few if any 145

horses with uncomplicated colic that could be relatively easily treated. In 146

addition, a relatively small number of samples were tested and further 147

investigation of LDH activity in abdominal fluid in horses with both complicated 148

and uncomplicated colic would be useful. 149

Another limitation of this study is the absence of a normal control population of 150

equine abdominal fluid. A previous study found that lactate dehydrogenase 151

activity in the peritoneal fluid of 20 normal horses was 143.0 ± 106.1 IU/L [17], 152

though surprisingly the method of LDH measurement was not mentioned. In our 153

study of 27 colic horses we calculated a mean of >550 (wet LDH L-P), >1000 (wet 154

LDH P-L) or >2500 (dry chemistry LDH P-L) IU/L LDH with 3 methods. Even the 155

lowest mean value is twice as high as LDH previously reported in peritoneal 156

effusions of healthy horses [17]. 157

The effect of hemolysis and blood contamination on LDH concentration were not 158

analysed in this study, however, cytological analysis of the equine effusions in 159

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

8

the non-survivors did not show PCVs or RBC content significantly different to the 160

survivors. Effects of hemolysis could be further investigated in larger study. 161

To our knowledge this is the first time that LDH activity in abdominal fluid from 162

horses with colic has been evaluated to differentiate between horses with colic 163

that survived the disease process for which they were admitted to the hospital 164

and those that did not survive (usually due to abdominal sepsis or neoplasia). 165

LDH activity in this population was not useful for differentiating between 166

whether horses required surgery or not (although LDH activity may be an 167

indicator of ischemia in humans [12]), nor between abdominal sepsis versus 168

neoplasia. LDH activity might therefore be a helpful test to indicate septic 169

peritonitis and neoplasia, whereas lactic acid is considered an important 170

indicator to assess bowel ischemia, though more research is needed. 171

5. Conclusions 172

LDH activity in abdominal effusion fluid may provide additional useful 173

prognostic information in horses with colic, with high values suggesting possible 174

neoplasia or sepsis as a cause for colic. LDH activity could not be used to 175

determine whether surgery was required or not and could not be used to 176

distinguish between sepsis and neoplasia as in both cases large number of LDH 177

containing, dead cells/bacteria are present in the effusion. LDH results varied 178

depending on the method of measurement used, therefore use of a consistent 179

method is highly recommended. There was moderate to good correlation 180

between lactate concentration and LDH activity in abdominal effusions from 181

colic horses and, although more research is needed, LDH activity might be an 182

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

9

interesting method to indicate septic peritonitis and neoplasia, whereas lactic 183

acid is an important indicator to assess intestinal ischemia [16, 18]. LDH might 184

have potential to function as an add on prognostic indicator in colic horses, 185

however, further comparison with peritoneal fluid lactic acid levels is required 186

to check for its added value both from a practical and scientific point of view. 187

Acknowledgements 188

Thanks to the MUVH equine clinical staff and the clinical pathology laboratory 189

staff for their help and support, and to Murdoch University who provided the 190

funding to make this project possible. 191

References 192

[1] Light RW. Parapneumonic effusions and empyema. Proc Am Thorac Soc. 193

2006;3:75-80. 194

[2] Light RW, Macgregor MI, Luchsinger PC, Ball WC, Jr. Pleural effusions: the 195

diagnostic separation of transudates and exudates. Ann Intern Med. 196

1972;77:507-13. 197

[3] Porcel JM, Light RW. Diagnostic approach to pleural effusion in adults. Am 198

Fam Physician. 2006;73:1211-120. 199

[4] Behera D. Pleural fluid lactate dehydrogenase and its isoenzymes in the 200

diagnosis of malignant and non-malignant pleural effusions. Lung Cancer. 201

2005;49:S140-S. 202

[5] Van Hoogmoed L, Rodger LD, Spier SJ, Gardner IA, Yarbrough TB, Snyder JR. 203

Evaluation of peritoneal fluid pH, glucose concentration, and lactate 204 dehydrogenase activity for detection of septic peritonitis in horses. J Am Vet Med 205

Assoc. 1999;214:1032-6. 206

[6] Zanatta R, Abate O, D'Angelo A, Miniscalco B, Mannelli A. Diagnostic and 207

prognostic value of serum lactate dehydrogenase (LDH) and LDH isoenzymes in 208 canine lymphoma. Vet Res Commun. 2003;27 Suppl 1:449-52. 209

[7] Funk D, Neiger R. [Usefulness of lactate dehydrogenase concentration in 210 thoracic and abdominal fluid of dogs]. Schweizer Archiv fur Tierheilkunde. 211

2014;156:577-83. 212

[8] Leise EM, Gray I, Ward MK. Leukocyte lactate dehydrogenase changes as an 213 indicator of infection prior to overt symptoms. J Bacteriol. 1968;96:154-9. 214 [9] Kaneko JJ, Harvey JW, Bruss ML. Clinical biochemistry of domestic animals. 215

6th ed. USA: Academic Press, Elsevier; 2008. p. 916. 216 [10] Panteghini M, Bais R. Enzymes. In: Burtis CA, Ashwood ER, Bruns DE, 217 editors. Tietz Fundamentals of clinical chemistry. USA: Saunders Elsevier; 2008. 218

p. 317-36. 219

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

10

[11] Drent M, Cobben NA, Henderson RF, Wouters EF, van Dieijen-Visser M. 220 Usefulness of lactate dehydrogenase and its isoenzymes as indicators of lung 221 damage or inflammation. Eur Respir J. 1996;9:1736-42. 222

[12] Rush BF, Jr., Host WR, Fewel J, Hsieh J. Intestinal ischemia and some organic 223 substances in serum and abdominal fluid. Arch Surg. 1972;105:151-7. 224 [13] Radcliffe RM, Buchanan BR, Cook VL, Divers TJ. The clinical value of whole 225 blood point-of-care biomarkers in large animal emergency and critical care 226

medicine. Journal of veterinary emergency and critical care. 2015;25:138-51. 227 [14] Tennent-Brown BS, Wilkins PA, Lindborg S, Russell G, Boston RC. 228 Assessment of a point-of-care lactate monitor in emergency admissions of adult 229 horses to a referral hospital. J Vet Intern Med. 2007;21:1090-8. 230

[15] Furr MO, Lessard P, White NA. Development of a colic severity score for 231 predicting the outcome of equine colic. Veterinary Surgery. 1995;24:97-101. 232 [16] Latson KM, Nieto JE, Beldomenico PM, Snyder JR. Evaluation of peritoneal 233 fluid lactate as a marker of intestinal ischaemia in equine colic. Equine Vet J. 234

2005;37:342-6. 235 [17] Brownlow MA, Hutchins DR, Johnston KG. Reference values for equine 236 peritoneal-fluid. Equine Vet J. 1981;13:127-30. 237

[18] van den Boom R, Butler CM, Sloet van Oldruitenborgh Oosterbaan MM. The 238

usability of peritoneal lactate concentration as a prognostic marker in horses 239

with severe colic admitted to a veterinary teaching hospital. Equine Vet Educ. 240

2010;22:420-5. 241 242

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Table 1: Table of 27 effusions from horses presenting with colic. TCC = total nucleated cell count, TP = total protein measured by refractometry, PCV = packed cell

volume, GIT = gastro-intestinal tract, ND = not done

Final Diagnosis TCC

(x109/L)

TP (g/L) PCV Lactate

(mmol/L)

Wet LDH L-P

(IU/L)

Wet LDH P-L

(IU/L)

Dry LDH P-L

(IU/L)

Survival >1 month

1 Colic, non-surgical 1.2 28 <0.01 1.15 124 220 660 Survived without surgery

2 Colic, non-surgical 2.1 <25 <0.03 ND 147 252 532 Survived without surgery

3 Colic, non-surgical 2.7 <25 <0.01 ND 71 129 528 Survived without surgery

4 Diarrhoea, hepatic fibrosis 0.9 32 <0.01 ND 132 267 571 Survived without surgery

5 Mild abdominal inflammation 1.4 <25 <0.01 ND 35 66 ND Survived without surgery

6 Peritonitis (no bacterial growth) 4.2 31 <0.01 ND 188 377 619 Survived without surgery

7 Peritonitis (no bacterial growth) 4.9 31 0.01 ND 197 392 601 Survived without surgery

8 Peritonitis (no bacterial growth) 3.2 30 0.01 ND 231 452 701 Survived without surgery

9 Septic peritonitis 8.5 35 0.02 3.6 281 497 771 Survived without surgery

10 Uterine haematoma/inflammation 39.1 39 0.01 ND 279 542 877 Survived without surgery

11 Colic, non-surgical, abdominal inflammation 60.2 41 <0.01 ND 1080 2815 6490 Survived without surgery

12 Cecal impaction, non-strangulating 180 degree cecal torsion 0.61 <25 <0.01 3 118 227 179 Survived with surgery

13 Colic (torsion, viable) 0.4 <25 <0.01 3.3 178 338 664 Survived with surgery

14 Colic strangulated colon, viable 2.6 <25 <0.01 5.1 264 476 899 Survived with surgery

15 Colon volvulus (black, viable) 0.2 <25 <0.01 2.1 153 281 ND Survived with surgery

16 Small colon feed impaction and a subacute grade 1 rectal tear 1.1 <25 <0.01 3.3 315 557 962 Survived with surgery

17 Splenic entrapment colic, no necrosis 1.1 26 <0.01 2.7 349 674 1092 Survived with surgery

18 Thromboembolic colic, ischemia of the large colon, resection 0.9 <25 <0.01 1.6 117 215 653 Survived with surgery

19 Uroperitoneum 0.1 <25 <0.01 ND 14 35 281 Survived with surgery

20 Duodenal rupture and sepsis 0.3 ND 0.03 13.6 416 773 2000 Died

21 Gastric squamous cell carcinoma 41.6 65 0.01 ND 2800 5380 15340 Died

22 Gastric squamous cell carcinoma 26.9 41 0.01 ND 376 658 995 Died

23 GIT necrosis and sepsis 0.65 54 0.04 ND 573 1092 1553 Died

24 GIT rupture and sepsis 31 35 0.02 ND 258 479 815 Died

25 Hepatic squamous cell carcinoma 32.3 33 0.01 ND 287 539 910 Died

26 Septic peritonitis 65.8 46 0.01 ND 1920 3380 11060 Died

27 Septic peritonitis (Actinobacillus sp.) 500 42 <0.01 ND 4920 8360 12300 Died

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPTTABLE 2: TABLE SHOWING CUT OFF VALUES OF LDH ACTIVITY IN ABDOMINAL EFFUSIONS ABOVE

WHICH THE PROGNOSIS IS POOR FOR SURVIVAL OF COLIC HORSES IN THIS STUDY

Methods of

measurement

LDH activity cut

off

Survival likely Survival unlikely Accuracy/area

under curve

Randox Wet

LDH L-P

280 IU/L Sensitivity 84%,

Specificity 87%

Sensitivity 87%,

Specificity 94%

0.85

Randox Wet

LDH P-L

500 IU/L Sensitivity 79%,

Specificity 87%

Sensitivity 87%,

Specificity 94%

0.81

IDEXX Dry

LDH P-L

900 IU/L Sensitivity 82%,

Specificity 87%

Sensitivity 87%,

Specificity 93%

0.84

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

1

1

2

3

y = 0.6036x - 17.705

r = 0.97 (p<0.01)

0

2000

4000

6000

8000

10000

12000

14000

0 5000 10000 15000 20000

Dif

fer

en

ce

be

twe

en

me

tho

ds

, IU

/L

Mean of both methods, IU/L

(b) Wet LDH L-P vs Dry LDH P-L

y = 0.2792x + 34.531

r = 0.98 (p<0.01)

0

500

1000

1500

2000

2500

3000

3500

4000

0 5000 10000 15000

Dif

fer

en

ce

be

twe

en

me

tho

ds

, IU

/L

Mean of both methods, IU/L

(a) Wet LDH L-P vs Wet LDH P-L

y = 0.3777x - 60.381

r = 0.91 (p<0.01)

0

2000

4000

6000

8000

10000

0 5000 10000 15000 20000

Dif

fer

en

ce

be

twe

en

me

tho

ds

, IU

/L

Mean of both methods, IU/L

(c) Wet LDH P-L vs Dry LDH P-L

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

2

FIGURE 1: BLAND AND ALTMAN PLOTS OF THE DATA OBTAINED FROM 27 EQUINE ABDOMINAL 4 EFFUSION SAMPLES ANALYSED COMPARING (A) WET LDH L-P TO WET LDH P-L (B) WET LDH L-P TO 5 DRY LDH P-L AND (C) WET LDH P-L TO DRY CHEMISTRY LDH P-L 6

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

FIGURE 2: COMPARISON OF LDH ACTIVITY MEANS (±SE) IN ABDOMINAL EFFUSIONS FROM HORSES

WITH NON-SURGICAL AND SURGICAL COLIC THAT SURVIVED AND THOSE THAT DIED WITH COLIC

SIGNS AND EFFUSIONS ASSOCIATED WITH SEPSIS OR NEOPLASIA THAT DIED, MEASURED BY 3

DIFFERENT METHODS

0

1000

2000

3000

4000

5000

6000

7000

8000

Wet LDH L-P Wet LDH P-L Dry LDH P-L

LD

H I

U/

L

Colic with survival

Sepsis/neoplasia

MANUSCRIP

T

ACCEPTED

ACCEPTED MANUSCRIPT

Highlights

LDH activity in abdominal fluid may aid in determining prognosis in horses with

colic. It has moderate to good correlation with lactate concentration and is

stable in refrigerated samples for at least 3 days. Different methods of

measuring LDH activity give markedly different values, and method of evaluation

must be known and consistent to provide useful information. Marked increases

in LDH activity in abdominal fluid are most consistent with a poor prognosis due

to advanced abdominal neoplasia and sepsis.