Effects of Aerobic and Anaerobic Fluid Collection on Biochemical Analysis of Peritoneal Fluid in Healthy Horses and Horses with

Colic
Alfredo E. Romero1, DVM, Jorge E. Nieto2, MVZ, PhD, Diplomate ACVS, Julie E. Dechant2, DVM, MS, Diplomate ACVS, Kate Hopper2, BVSc, PhD, Diplomate ACVECC, and Monica Aleman3, MVZ, PhD, Diplomate ACVIM
1

William R. Pritchard Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California-Davis, Davis, CA, 2Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California-Davis, Davis, CA and 3Department of Medicine and Epidemiology, School of Veterinary Medicine, University of California-Davis, Davis, CA

Corresponding Author Jorge E. Nieto, Department of Surgical and Radiological Sciences, School of Veterinary Medicine, University of California-Davis, Davis, CA 95616 E-mail: jenieto@ucdavis.edu Submitted December 2008 Accepted January 2010 DOI:10.1111/j.1532-950X.2010.00767.x

Objective: To determine whether in healthy horses and those with colic, exposure of peritoneal uid to room air affects values obtained on biochemical analysis. Study Design: Prospective study. Animals: Adult horses with a primary complaint of acute abdominal pain (n = 29) and 12 healthy horses. Methods: Peritoneal uid was aseptically collected under aerobic and anaerobic conditions. After collection, pH, PCO2, PO2, HCO, Na1, ionized Ca21, K1, lac3 tate, and glucose were immediately measured using a commercial blood gas analyzer. Biochemical variables were compared between aerobically and anaerobically obtained samples using a paired t-test. Results: In healthy horses, peritoneal uid samples collected under anaerobic conditions had higher PCO2 and ionized Ca21 and lower PO2, HCO, and pH 3 compared with samples exposed to air. No differences were observed for K1, Na1, glucose, and lactate. In horses with colic, samples collected anaerobically had higher PCO2, ionized Ca21, Na1, and glucose and lower PO2, HCO, and pH 3 value compared with samples exposed to air. No differences were observed for K1 and lactate. Conclusion: Exposure of peritoneal uid to room air had a significant effect on pH, PCO2, PO2, and variables associated or dependent on changes in pH such as HCO and ionized Ca21. Interpretation of biochemical analysis of peritoneal uid 3 may be inuenced by sample collection method.

Abdominocentesis for collection of peritoneal uid in horses was reported by Maksic1 to aid in the diagnosis and prognosis of colic, dystocia, and inammatory and noninammatory conditions.The ideal peritoneal uid marker for the diagnosis of specic lesions has not been identied. At best, certain markers show an increased sensitivity for the presence of intestinal ischemia but are nonspecic, and are more helpful when used in combination with other laboratory values and physical examination ndings. Variables routinely evaluated in peritoneal uid samples include volume and color, total white cell count and differential, and total protein concentration.2 Other studies have described additional biochemical variables that could be used as biomarkers of sepsis, ischemia, need of surgical intervention, and survival.36 In horses, peritoneal uid pH and glucose concentration are lower with septic peritonitis.6 Peritoneal lactate concentration can be evaluated using bench top and portable analyzers 3,4,79

and a blood gas analyzer has been used to evaluate peritoneal uid variables that correlate with intestinal ischemia.4 For strangulating obstruction, peritoneal uid variables with the strongest correlation included alterations in gross appearance (color), chloride, pH, and lactate.4 Evaluation of glucose concentration and pH are key components in the diagnosis of pleural effusion type in people; however, exposure of pleural uid to air signicantly increased pH values, which in turn may inuence clinical management.10,11 Pleural uid pH was also significantly decreased by residual lidocaine and heparin in the collection syringe.10 Previous studies in horses have used peritoneal uid obtained by a free catch sampling method. We are unaware that the inuence of exposure to room air on biochemical variables in equine peritoneal uid has been investigated. Thus our purpose was to determine whether exposure to room air alters values obtained on biochemical analysis of peritoneal uid from healthy

40

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

Romero et al.

Equine Peritoneal Fluid Analysis

horses and horses with colic. Our hypothesis was that exposure of peritoneal uid to room air would alter results of biochemical analysis compared with results obtained from peritoneal uid samples collected anaerobically.

MATERIALS AND METHODS

Preliminary Studies To eliminate causes of variation other than exposure to air, we designed a series of experiments to determine whether the measured analytes vary in sequential peritoneal uid samples (healthy group A), and to determine whether different forms of heparin affect results (healthy group B). We also evaluated in healthy horses from group B, if different peritoneal uid collection methods (aerobic and anaerobic) affected these biochemical variables. Horses Healthy Horses. Twelve clinically healthy horses from our research herd were subdivided into 2 groups (groups A, B) of 6 horses each. Horses with Colic. Horses, admitted over 7 months, with acute abdominal pain were studied. Inclusion criteria were complete medical records, analysis of peritoneal uid, and collection by 2 authors (AR, JN) to avoid variations in peritoneal uid collection methodology. Diagnostic evaluation on admission included physical examination, rectal palpation, nasogastric intubation, abdominal radiography and ultrasonography, venous blood evaluation (leukocyte count, packed cell volume, total protein concentration, blood gases, serum biochemical panel, and lactate concentration), and abdominocentesis. Abdominocentesis A 37 cm2 area to the right of midline, at the most dependent region of the ventral abdomen, was clipped and aseptically prepared with alcohol and povidoneiodine solution. Lidocaine hydrochloride (12 mL) was inltrated subcutaneously to provide local anesthesia before the nal aseptic preparation. A stab incision (#15 scalpel blade) was made through skin and external sheath of the rectus abdominis muscle, then a sterile bitch catheter connected to a 10 cm silicon tube and a 3-way stopcock was advanced through the incision into the peritoneal cavity to collect peritoneal uid. After collection, the catheter was removed. The stab incision was not closed. Peritoneal Fluid Collection Healthy Horses. Group A (n = 6) had 3 consecutive samples of peritoneal uid collected under aerobic conditions to determine whether sequential sampling affects analyzed biochemical variables. Peritoneal uid was allowed to ow freely from the collection device for exposure to room air and the 3 samples (2 mL each) were collected sequentially,

into 3 mL Vacutainer tubes containing lithium heparin (51 USP units of dry lithium heparin; Becton Dickinson, Franklin Lakes, NJ). Group B (n = 6) had 2 samples collected under aerobic conditions (free ow) to determine whether different heparin salts can affect biochemical variables. One specimen was 2 mL peritoneal uid collected into a 3 mL Vacutainer tube containing 51 USP units of dry lithium heparin (Becton Dickinson) and the 2nd sample was 2 mL peritoneal uid collected into a 3 mL Vacutainer tube containing 51 USP units of sodium heparin (volume of 10.2 mL; Becton Dickinson). Specimens were collected sequentially with no time interval between samples. A 3rd peritoneal uid sample was collected anaerobically from group B horses to determine whether biochemical outcome variables were inuenced by collection method (aerobic, anaerobic). A 3 mL syringe was heparinized by adding 51 USP units of sodium heparin 1:5000 (total volume 10.2 mL) and then attached to a lateral port of the stopcock on the catheter system. Once peritoneal uid lled the silicon tube, removing the air, the stopcock was switched to allow collection of 2 mL peritoneal uid with the syringe. The syringe was disconnected and immediately airtight sealed using a syringe tip cap (Jansen Medical LLC., Houston, TX). Fluid obtained by aerobic (lithium heparin) and anaerobic methods were compared to determine whether differences in analyte values occur in response to room air exposure. Horses with Colic. Peritoneal uid was collected anaerobically and aerobically from each horse as described for healthy horses. Fluid was analyzed and measured values of analytes compared between collection methods. Peritoneal Fluid Analysis Specimens were analyzed (ABL 700; Radiometer America Inc., Westlake, OH) within 5 minutes of collection. Twice daily, the analyzer was calibrated by a technician using the manufacturers instructions, and autocalibration occurred 4-hourly. The measured variables were pH, PCO2, PO2, HCO, K1, Na1, ionized Ca21, Cl, glucose, and lactate 3 concentrations. Statistical Analysis Statistical analysis was performed for the variables: pH, PCO2, PO2, HCO, K1, Na1, ionized Ca21, Cl, glucose, 3 and lactate concentration using commercially available software (SPSS version 12.0; SPSS Inc., Chicago, IL). Biochemical variables obtained from the 3 serial samples in group A were analyzed by repeated measures ANOVA. Comparison between samples collected with sodium and lithium heparin in group B was performed using a 2-tailed paired t-test. Variables obtained from aerobic and anaerobic collection methods in healthy and diseased horses were also compared using a 2-tailed paired t-test. Significance was set at P o .05.

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

41

Equine Peritoneal Fluid Analysis

Romero et al.

RESULTS
Preliminary Studies Serial Control Samples. There was no significant difference in measured individual analyte values between 3 consecutive samples of peritoneal uid exposed to air from healthy horses (group A). Effect of Heparin Salt Type. Peritoneal uid samples (group B healthy horses) exposed to air and collected in lithium heparin had a lower Cl concentration (104.5 mEq/ L) compared with those collected in sodium heparin (105.8 mEq/L; P = .025). Significant differences were not identied for other analytes. Aerobic Versus Anaerobic Collection pH, PCO2, PO2, HCO, and ionized Ca21 values were sig3 nificantly different between peritoneal uid samples (healthy horses, group B) collected under aerobic (lithium heparin) and anaerobic (sodium heparin) conditions (Table 1) whereas no differences were observed for K1, Na1, glucose, and lactate concentrations. Because of the observed differences in Cl concentration when using different heparin salts, Cl was excluded for comparison of these collection methods. Horses with Colic Twenty-nine horses met the inclusion criteria. Clinical diagnoses were 9 nonstrangulating colonic obstructions (nephrosplenic entrapment [n = 3], large colon impaction [2], small colon impaction [2], and enterolith [2]); 6 strangulat-

ing small intestinal obstructions (pedunculated lipoma [4], epiploic foramen entrapment [1], gastrosplenic rent [1]); 2 nonstrangulating small intestinal obstructions (ileal impaction [1] and proximal duodenitis/jejunitis [1]); 3 horses with peritonitis (1 with duodenal perforation and a mixed bacterial growth, 1with intracellular cocci identied on Gram stain, 1 with a nonseptic peritonitis); 1 with gastric rupture; 1 with an eosinophilic lymphoplasmacytic enterocolitis; and 7 with medical colic of undetermined cause. Statistical differences between aerobic and anaerobic collection methods were found for pH, PCO2, PO2, HCO, 3 Na1, ionized Ca21, and glucose (Table 1). No differences were observed for K1, and lactate concentration (Table 1).

DISCUSSION
Accurate interpretation of peritoneal uid biochemical variables is important for the appropriate management of clinical cases. We found statistical differences for various biochemical analytes (PCO2, PO2, HCO, pH, Na1, ionized 3 Ca21, and glucose concentration) between peritoneal uid samples collected by aerobic and anaerobic methods. Samples collected aerobically had a mean increase in pH of 0.17 and 0.16 U in healthy and diseased horses, respectively. This is similar to the increase in pH (0.16) observed in aerobically collected pleural uid in people.11 In our study, two-thirds of the Vacutainer tubes were lled with peritoneal uid, allowing air to be in contact with the surface of the sample. It is uncertain if lling the tube completely would have produced a smaller change in biochemical values by limiting the exposure to air and gas exchange. Peritoneal uid was analyzed within 5 minutes of collection to prevent alterations that may occur during

Table 1 Mean SD, Mean Differences and P-Values for Peritoneal Fluid Collected Aerobically and Anaerobically from 6 Healthy and 29 Horses with Colic Anaerobic Healthy horses PCO2(mmHg) PO2(mmHg) HCO (mmol/L) 3 pH Potassium (mEq/L) Sodium (mEq/L) Ionized Ca21 (mmol/L) Glucose (mg/dL) Lactate (mmol/L) Horses with colic PCO2 (mmHg) PO2 (mmHg) HCO (mmol/L) 3 pH Potassium (mEq/L) Sodium (mEq/L) Ionized Ca21 (mmol/L) Glucose (mg/dL) Lactate (mmol/L) 43.15 11.3 108.85 21.1 33.07 4.3 7.51 0.1 3.88 0.3 135.67 1 1.51 0.01 113.83 5.2 0.7 0.5 52.69 15.4 82.89 32.4 31.30 6.1 7.4 0.2 3.56 0.6 134.3 3.8 1.33 0.2 112.69 43.3 3.46 3.9 Aerobic 33.05 11.3 149.83 17.7 35.52 4.2 7.67 0.2 3.83 0.3 135.67 1.6 1.37 0.2 114.5 7.4 0.68 0.4 39.78 12.1 163.65 34.7 35.16 8.0 7.56 0.2 3.59 0.5 133.21 2.4 1.25 0.2 117.34 43.9 3.54 4 Difference 10.1 40.98 0.45 0.17 0.05 0 0.14 0.67 0.02 12.4 80.76 3.86 0.16 0.03 1.09 0.08 4.65 0.08 P-Value .011 .008 .009 .027 .363 1.0 .045 .699 .695 o .001 o .001 o .001 o .001 .458 .012 .001 .001 .076

Significant differences in analyte values between collection methods. Significance was set at P o .05.

42

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

Romero et al.

Equine Peritoneal Fluid Analysis

storage. It is unknown if storing the sample on ice before analysis would decrease changes in measured variables. For human pleural uid, no differences in pH were observed if samples were stored on ice or at 371C for 30 minutes.11 Further, pleural uid pH was stable at room temperature for 1 hour and significantly increased at 4 and 24 hours,10 suggesting that the time from sample collection to its analysis is more important than the method of collection and storage. However, the method of storage may inuence the time available for an accurate analysis and interpretation of results. It was suggested that a pH difference of Z.05 for pleural uid should be considered as the threshold for clinical significance10; however, it is unknown if the same variation would apply to equine peritoneal uid. Lidocaine, administered as a local anesthetic before pleural aspiration in people results in a reduction in pleural uid pH.12 A small amount of residual lidocaine (0.2 mL) causes a clinically significant drop in pH that worsens in a dose-dependent manner.10 Therefore clinicians need to be aware of this limiting the amount of lidocaine used, and using a different needle and syringe for sample collection. Diagnosis of septic peritonitis is based on physical examination observations, hematologic ndings, and peritoneal uid analysis.13 The diagnosis of septic peritonitis after abdominal surgery is complicated because the peritoneal uid total nucleated cell count and protein concentration are elevated as a consequence of surgery.14 Typically, identication of bacteria on cytology or culture is a more denitive indicator of septic peritonitis; bacteria are not always observed on cytology or grow on culture plates and bacterial growth may take days. In a clinical study, horses with septic peritonitis had a lower peritoneal uid pH than horses with nonseptic peritonitis6; negative predictive value was 91% and positive predictive value was 100% with a cutoff pH of o 7.3. Therefore, slight changes in pH may be relevant clinically and the method of collection should be considered when using peritoneal uid pH for the diagnosis of septic peritonitis. Using pH values from an anaerobically collected sample of peritoneal uid may guide the clinician toward an incorrect diagnosis of septic peritonitis. Examination of peritoneal uid is a common diagnostic tool in horses with colic. Evaluation of peritoneal uid color, total protein concentration, and total white cell count are the variables more commonly used clinically; however, recently other biochemical variables have been evaluated in horses with colic.4,68 Horses with colic and horses with strangulating lesions or ruptured viscus have lower peritoneal uid pH than healthy horses or horses with nonstrangulating lesions or medical colic.4,8 Alterations in peritoneal uid gross appearance, and pH, chloride, and lactate concentrations had the strongest correlation with presence of intestinal ischemia from strangulating obstruction.4 Ischemia decreased pH in the uid surrounding segments of strangulated bowel in ponies.15 Thus the potential effect of changes in pH by the method of peritoneal uid collection in horses with intestinal ischemia remains to be determined.

Peritoneal lactate concentration is a valuable diagnostic aid in horses with colic.4,7,8 In an earlier study using the same analyzer, we found that horses with an intestinal strangulating obstruction had a higher peritoneal lactate concentration (8.45 mmol/L; reference value o 2.0 mmol/ L) than those with a nonstrangulating obstruction (2.09 mmol/L).4 We found no effect of air exposure on peritoneal lactate concentration in healthy horses or horses with colic. Ionized Ca21 concentrations were significantly lower in peritoneal uid collected aerobically compared with anaerobic collection, both in healthy horses and those with colic. As noted, peritoneal uid pH was higher in uid collected aerobically. It is known that hydrogen ions compete with calcium for binding sites on albumin and other proteins.16 Therefore, decreases in hydrogen ions (increased pH) will result in less free calcium (ionized Ca21). The rate of change of ionized Ca21 concentration with pH change is of 0.36 mmol/L per pH unit in plasma samples from people.16 Glucose concentration was higher in aerobic samples from horses with colic. In people, presence of air in pleural uid samples had no effect on glucose values10; differences 4 18 mg/dL were considered clinically relevant. In horses, glucose concentration has been identied as a marker for septic peritonitis6; serum-to-peritoneal uid glucose concentration differences 4 50 mg/dL and an abdominal glucose concentration of o 30 mg/dL were indicative of septic peritonitis. Therefore, it is unlikely that the observed difference in glucose concentration of 4.65 mg/dL between aerobic and anaerobic collection of peritoneal uid would impact the diagnosis in diseased horses. When peritoneal uid samples were collected aerobically, there was a significant decrease in PCO2 and a significant increase in pH, bicarbonate, and PO2 values, compared with samples collected under anaerobic conditions. Changes in PCO2 and PO2 upon exposure to air can be attributed to gas equilibration. The increase in pH is expected with the decrease in PCO2 but the increase in bicarbonate concentration was unexpected. Blood gas analyzers measure pH and PCO2 and use these values to calculate the bicarbonate concentration from the HendersonHasselbach equation. The significant increase in bicarbonate concentration in the aerobic samples reects a greater increase in measured pH than the measured decrease in PCO2 can account for. The cause of this high pH (and subsequent high calculated bicarbonate concentration) is not easily explained. An increase in pleural uid pH after air exposure has been reported10 and it was suggested that the change in pH was because of loss of PCO2 from the sample but unfortunately PCO2 levels were not measured so the true mechanisms of the pH change could not be established.10 As aerobic collection of uid samples is not expected to alter pH beyond that related to changes in PCO2, it suggests that the different method of sample collection for aerobic and anaerobic samples may have altered pH. Aerobic samples were collected in a Vacutainer containing dry lithium

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

43

Equine Peritoneal Fluid Analysis

Romero et al.

heparin whereas anaerobic samples were collected using a syringe containing liquid sodium heparin. Because liquid sodium heparin is relatively acidic (pH $6.655), its addition to anaerobic samples could account for the lower pH (beyond that accounted for by changes in PCO2) compared with the aerobic samples. But the results from the group B horses in our study which compared peritoneal uid sample analysis using dry lithium heparin with an equivalent dose of liquid sodium heparin demonstrated no significant difference in pH between the 2 anticoagulant methods. The cause of the increase in pH beyond that predicted by the change in PCO2, and hence the increased calculated bicarbonate concentration of the aerobic samples in our study cannot be readily explained but emphasizes the necessity for maintaining a consistent protocol for peritoneal uid collection to avoid variation associated with sample collection technique. Because more than 1 uid collection method was going to be performed in all horses in our study, it was important to investigate if sequential collections would have an effect on measured biochemical variables. Statistical differences in the analytes studied were not identied, indicating that each sample was representative of the peritoneal uid. Therefore, any observed differences in variables studied were the result of the different collection methods and not sequential sampling. It was determined that sequential uid collections did not significantly affect the values of various analytes. The effects of collecting peritoneal uid into different heparin salts on various biochemical variables have not been studied in horses. Other than Cl concentrations, significant differences in measurable analytes were not observed using different heparin salts at the concentrations used for peritoneal uid collection in this study. Sodium and lithium heparin bind to calcium and reduce the ionized Ca21 levels in blood.17 This was not evaluated in our study. Sodium heparin did not alter the Na1 concentrations in peritoneal uid when compared with samples collected in lithium heparin. A limitation of our study was using a liquid and a powder source of anticoagulant as sample dilution does create a preanalytical error.17,18 This error can occur when the ratio of heparin to sample is high or heparin is not dried17,18; however, this was minimized in our study using a small amount of liquid heparin (0.5% of total volume). In human pleural uid, residual sodium heparin (13% of total sample) decreased pH values, compared with samples where heparin was expelled from the syringe.10 In people and dogs, using heparin salt at o 45% of the total sample volume minimizes this effect.17,18 With the exception of Cl concentrations, there were no statistical differences in peritoneal uid electrolytes, gases, lactate, or glucose concentrations. From the variables that were statistically different depending on collection method (aerobic, anaerobic), the ones that had larger variability were pH, PCO2, PO2, and HCO. Differences in results were likely mediated by equi3 libration of the aerobic sample with atmospheric CO2 and O2. If pH is used in evaluation of clinical cases especially

those with septic peritonitis, clinicians need to be aware that the method of peritoneal uid collection can inuence interpretation of the results. If anaerobic collection of peritoneal uid becomes routine clinically, determination of reference intervals for pH is warranted.

ACKNOWLEDGMENTS
We thank Dr. Philip Kass for statistical advice.

REFERENCES
1. Maksic D: Abdominal paracentesis and its use in diagnosis in the horse. Proc Am Assoc Equine Practnr 1964;20: 319324 2. Swanwick RA, Wilkinson JS: A clinical evaluation of abdominal paracentesis in the horse. Aust Vet J 1976;52:109117 3. Parry BW, Anderson GA, Gay CC: Prognosis in equine colic: a study of individual variables used in case assessment. Equine Vet J 1983;15:337344 4. Latson KM, Nieto JE, Beldomenico RM, et al: Evaluation of peritoneal uid lactate as a marker of intestinal ischaemia in equine colic. Equine Vet J 2005;37:342346 5. Nieto JE, Aldridge BM, Beldomenico PM, et al: Characterization of equine intestinal fatty acid binding protein and its use in managing horses with colic. Am J Vet Res 2005;66:223232 6. Van Hoogmoed L, Rodger LD, Spier SJ, et al: Evaluation of peritoneal uid pH, glucose concentration, and lactate dehydrogenase activity for detection of septic peritonitis in horses. J Am Vet Med Assoc 1999;214:10321036 7. Delesalle C, Dewulf J, Lefebvre RA, et al: Determination of lactate concentrations in blood plasma and peritoneal uid in horses with colic by an Accusport analyzer. J Vet Intern Med 2007;21:293301 8. Saulez MN, Cebra CK, Dailey M: Comparative biochemical analyses of venous blood and peritoneal uid from horses with colic using a portable analyser and an in-house analyser. Vet Rec 2005;157:217223 9. Sloet van Oldruitenborgh-Oosterbaan MM, van den Broek ET, Spierenburg AJ: Evaluation of the usefulness of the portable device Lactate Pro for measurement of lactate concentrations in equine whole blood. J Vet Diagn Invest 2008;20:8385 10. Rahman NM, Mishra EK, Davies HE, et al: Clinically important factors inuencing the diagnostic measurement of pleural uid pH and glucose. Am J Respir Crit Care Med 2008;178:483490 11. Venkatesh B, Boots RJ, Wallis SC: Accuracy of pleural uid pH and PCO2 measurement in a blood gas analyser. Analysis of bias and precision. Scand J Clin Lab Invest 1999;59:619626 12. Jimenez Castro D, Diaz G, Perez-Rodriguez E, et al: Modication of pleural uid pH by local anesthesia. Chest 1999;116:399402

44

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

Romero et al.

Equine Peritoneal Fluid Analysis

13. Markel MD: Prevention and management of peritonitis in horses. Vet Clin North Am Equine Pract 1988;4: 145156 14. Hanson RR, Nixon AJ, Gronwall R, et al: Evaluation of peritoneal uid following intestinal resection and anastomosis in horses. Am J Vet Res 1992;53:216221 15. Ruggles AJ, Freeman DE, Acland HM, et al: Changes in uid composition on the serosal surface of jejunum and small colon subjected to venous strangulation obstruction in ponies. Am J Vet Res 1993;54:333340

16. Wang S, McDonnell EH, Sedor FA, et al: pH effects on measurements of ionized calcium and ionized magnesium in blood. Arch Pathol Lab Med 2002;126:947950 17. Higgins C: The use of heparin in preparing samples for bloodgas analysis. Med Lab Obs 2007;39:1618, 20, quiz 2213 18. Hopper K, Rezende ML, Haskins SC: Assessment of the effect of dilution of blood samples with sodium heparin on blood gas, electrolyte, and lactate measurements in dogs. Am J Vet Res 2005;66:656660

c Veterinary Surgery 40 (2011) 4045  Copyright 2010 by The American College of Veterinary Surgeons

45