• View in gallery
    Figure 1—

    Diagram of an apparatus for determination of transepicardial flow in dogs. A hemispheric plastic capsule is attached to the left ventricular epicardial surface and filled with an albumin-saline (0.9% NaCl) solution. Fluid-filled tubing extending from the capsule allows measurement of intracapsular hydrostatic pressure (PP), which represents pericardial hydrostatic pressure, and flow into or out of the capsule (QP), which represents transepicardial flow (JV,e).

  • View in gallery
    Figure 2—

    Effect of hydrostatic pressure in the epicardial capsule on transepicardial flow of capsule solutions 1 (closed circles and solid line; COP = 7.0 mm Hg) and 2 (open circles and dashed line; COP = 2.0 mm Hg) in 8 heartworm-free (A) and 7 heartworm-infected (B) dogs. (Data in part A adapted from Stewart RH, Rohn DA, Allen SJ, et al. Basic determinants of epicardial transudation. Am J Physiol Heart Circ Physiol 1997;273:H1408–H1414. Reprinted with permission from the American Physiological Society.)

  • View in gallery
    Figure 3—

    The effect of heartworm status on interstitial driving pressure and myocardial water content. Interstitial driving pressure (Pint − σd,e0int) is plotted as a function of myocardial water-todry weight ratio for heartworm-infected dogs (black circle) and heartworm-free dogs (white circle). A regression line (dashed line) derived from Laine and Granger14 has been added to indicate the expected relationship between myocardial interstitial hydrostatic pressure and water content. Data are expressed as mean ± SEM.

  • 1.

    Laine GA, Allen SJ. Left ventricular myocardial edema: lymph flow, interstitial fibrosis, and cardiac function. Circ Res 1991;68:17131721.

  • 2.

    Davis KL, Mehlhorn U, Laine GA, et al. Myocardial edema, left ventricular function, and pulmonary hypertension. J Appl Physiol 1995;78:132137.

  • 3.

    Mehlhorn U, Davis KL, Burke EJ, et al. Impact of cardiopulmonary bypass and cardioplegic arrest on myocardial lymphatic function. Am J Physiol 1995;268:H178–H183.

    • Search Google Scholar
    • Export Citation
  • 4.

    Pratt JW, Schertel ER, Schaefer SL, et al. Acute transient coronary sinus hypertension impairs left ventricular function and induces myocardial edema. Am J Physiol 1996;271:H834–H841.

    • Search Google Scholar
    • Export Citation
  • 5.

    Miyamoto M, McClure DE, Schertel ER, et al. Effects of hypoproteinemia-induced myocardial edema on left ventricular function. Am J Physiol 1998;274:H937–H944.

    • Search Google Scholar
    • Export Citation
  • 6.

    Rohn DA, Davis KL, Mehlhorn U, et al. Myocardial edema and compromised left ventricular function attributable to dirofilariasis and cardiopulmonary bypass in dogs. Am J Vet Res 1995;56:221226.

    • Search Google Scholar
    • Export Citation
  • 7.

    Stewart RH, Rohn DA, Allen SJ, et al. Basic determinants of epicardial transudation. Am J Physiol Heart Circ Physiol 1997;273: H1408–H1414.

    • Search Google Scholar
    • Export Citation
  • 8.

    Michel CC. Fluid movements through capillary walls. In: Renkin EM, Michel CC, eds. Handbook of physiology, section 2: the cardiovascular system. Vol IV. New York: Oxford University Press, 1984;375409.

    • Search Google Scholar
    • Export Citation
  • 9.

    Negrini D, Townsley MI, Taylor AE. Hydraulic conductivity of canine parietal pleura in vivo. J Appl Physiol 1990;69:438442.

  • 10.

    Mason RL, Gunst RF, Hess JL. Analysis of covariance for completely randomized designs. In: Statistical design and analysis of experiments: with applications to engineering and science. New York: Wiley, 1989;398406.

    • Search Google Scholar
    • Export Citation
  • 11.

    Mehlhorn U, Allen SJ, Adams D, et al. Normothermic continuous antegrade blood cardioplegia does not prevent myocardial edema and cardiac dysfunction. Circulation 1995;92:19391946.

    • Search Google Scholar
    • Export Citation
  • 12.

    Gibson AT, Segal MB. A study of the composition of pericardial fluid, with special reference to the probable mechanism of fluid formation. J Physiol 1978;277:367377.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13.

    Maurer FW, Warren MF, Drinker CK. The composition of mammalian pericardial and peritoneal fluids. Am J Physiol 1940;128:635644.

  • 14.

    Laine GA, Granger HJ. Microvascular, interstitial and lymphatic interactions in normal heart. Am J Physiol 1985;249:H834–H842.

  • 15.

    Reed RK, Rodt SA. Increased negativity of interstitial fluid pressure during the onset stage on inflammatory edema in rat skin. Am J Physiol 1991;260:H1985–H1991.

    • Search Google Scholar
    • Export Citation
  • 16.

    Wiig H, Rubin K, Reed RK. New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences. Acta Anaesthesiol Scand 2003;47:111121.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Grinnell F, Lamke CR. Reorganization of hydrated collagen lattices by human skin fibroblasts. J Cell Sci 1984;66:5163.

  • 18.

    Berg A, Ekwall A-KH, Rubin K, et al. Effect of PGE1, PGI2, and PGF2 A analogs on collagen gel compaction in vitro and interstitial pressure in vivo. Am J Physiol 1998;274:H663–H671.

    • Search Google Scholar
    • Export Citation

Advertisement

Myocardial fluid balance in dogs with naturally acquired heartworm infection

Randolph H. StewartMichael E. DeBakey Institute, Texas A&M University, College Station, TX 77843; and the Center for Microvascular and Lymphatic Studies, The University of Texas Medical School, Houston, TX 77030.

Search for other papers by Randolph H. Stewart in
Current site
Google Scholar
PubMed
Close
 DVM, PhD
,
Karen UrayMichael E. DeBakey Institute, Texas A&M University, College Station, TX 77843; and the Center for Microvascular and Lymphatic Studies, The University of Texas Medical School, Houston, TX 77030.

Search for other papers by Karen Uray in
Current site
Google Scholar
PubMed
Close
 PhD
,
Charles S. CoxMichael E. DeBakey Institute, Texas A&M University, College Station, TX 77843; and the Center for Microvascular and Lymphatic Studies, The University of Texas Medical School, Houston, TX 77030.

Search for other papers by Charles S. Cox in
Current site
Google Scholar
PubMed
Close
 MD
, and
Glen A. LaineMichael E. DeBakey Institute, Texas A&M University, College Station, TX 77843; and the Center for Microvascular and Lymphatic Studies, The University of Texas Medical School, Houston, TX 77030.

Search for other papers by Glen A. Laine in
Current site
Google Scholar
PubMed
Close
 PhD

Abstract

Objective—To determine the effect of naturally acquired heartworm (Dirofilaria immitis) infection on myocardial fluid balance as indicated by myocardial water content and the dynamics of transepicardial fluid flow.

Animals—7 dogs infected with adult heartworms and 8 dogs free of heartworm infection.

Procedures—Infected dogs had heartworms in the right ventricle, pulmonary artery, or both but no evidence of cardiovascular disease on physical examination. A hemispheric capsule was attached to the epicardial surface of all dogs for determination of transepicardial fluid dynamics and permeability of the epicardium to water and protein. Myocardial water content and hydroxyproline content were assessed at necropsy.

Results—Myocardial water content was significantly lower in heartworm-infected dogs. No differences in myocardial hydroxyproline content, transepicardial fluid flow, or epicardial water or protein permeability were detected.

Conclusions and Clinical Relevance—Heartworm infection significantly altered myocardial fluid balance in dogs, possibly because of a change in the myocardial interstitial pressure-volume relationship. These changes may be associated with increased vulnerability to cardiovascular stressors in heartworm-infected dogs.

Abstract

Objective—To determine the effect of naturally acquired heartworm (Dirofilaria immitis) infection on myocardial fluid balance as indicated by myocardial water content and the dynamics of transepicardial fluid flow.

Animals—7 dogs infected with adult heartworms and 8 dogs free of heartworm infection.

Procedures—Infected dogs had heartworms in the right ventricle, pulmonary artery, or both but no evidence of cardiovascular disease on physical examination. A hemispheric capsule was attached to the epicardial surface of all dogs for determination of transepicardial fluid dynamics and permeability of the epicardium to water and protein. Myocardial water content and hydroxyproline content were assessed at necropsy.

Results—Myocardial water content was significantly lower in heartworm-infected dogs. No differences in myocardial hydroxyproline content, transepicardial fluid flow, or epicardial water or protein permeability were detected.

Conclusions and Clinical Relevance—Heartworm infection significantly altered myocardial fluid balance in dogs, possibly because of a change in the myocardial interstitial pressure-volume relationship. These changes may be associated with increased vulnerability to cardiovascular stressors in heartworm-infected dogs.

The importance of understanding the mechanisms regulating myocardial fluid balance has been emphasized by determination that interstitial edema formation in the myocardium leads to systolic and diastolic dysfunction.1–5 Myocardial interstitial edema may be caused by right ventricular hypertension, coronary venous hypertension, arterial hypertension, cardiopulmonary bypass with cardioplegic arrest, and hypoproteinemia.1–5 In a previous study,6 dogs with naturally acquired heartworm (Dirofilaria immitis) infection had significantly higher myocardial water content than uninfected dogs and were more likely than uninfected dogs to develop fatal cardiac dysfunction when challenged with cardiopulmonary bypass. These findings prompted us to further investigate the effects of heartworm infection on myocardial interstitial fluid balance.

In addition to measurement of myocardial water-todry weight ratio, a technique used to assess myocardial fluid balance has been developed that involves measurement of the fluid flow across the epicardium from the myocardial interstitial space.7 By use of the Starling-Landis equation as a model, the forces governing this transepicardial flow have been expressed as follows:

article image
where JV,e is the rate of transepicardial flow; Lp,e is the epicardial hydraulic conductance; Ae is epicardial surface area; Pint and Pp are the hydrostatic pressures within the myocardial interstitium and the fluidfilled pericardial space, respectively; σd,e is the osmotic reflection coefficient of the epicardium with a value between 0 and 1; and 0int and 0p are the COPs exerted by the interstitial and pericardial fluids, respectively (equation 1).7

The LP,e represents the permeability of the epicardial barrier to water. Transepicardial flux (JV,e/Ae), flow per unit surface area, changes linearly as a function of Pp.7 The Lp,e can be determined from the slope of this relationship7 by use of equation 2:

article image

The σd,e, an index of protein permeability, is a measure of the effectiveness with which the COP gradient is expressed as a hydrostatic pressure gradient across a semipermeable membrane, in this case, the epicardium.8 The value of σd,e equals 0 when the barrier is freely permeable to protein and equals 1 when the barrier is completely impermeable to protein. Thus, σd,e can be estimated by the following equation:

article image
where b1-2 is the change in transepicardial flux that results from changing the pericardial fluid COP from 0p,1 to 0p,2 (equation 3).7 For this purpose, a hemispheric capsule is affixed to the epicardial surface so that the hydrostatic pressure and COP of the solution within the capsule represent Pp and 0p, respectively; fluid flow across the epicardium into and out of the capsule represents JV,e, and the epicardial surface area enclosed by the capsule represents Ae.7

The principal objective of the study reported here was to further characterize the effects of naturally acquired heartworm infection on myocardial interstitial fluid balance and transepicardial fluid flow.

Materials and Methods

All procedures were conducted following approval by the Texas A&M University Laboratory Animal Care Committee and were consistent with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Animal preparation and instrumentation—Adult dogs (n = 15) with body mass > 15 kg and of either sex were used. The dogs were obtained from a population known to have a high prevalence of naturally acquired heartworm infection. No dogs had evidence of cardiovascular disease on close physical examination. Investigators were unaware of the state of heartworm infection. The dogs were assigned to 1 of 2 groups, heartworm infected or heartworm free, at the end of each experiment, according to the presence or absence of adult heartworms in the right ventricle, pulmonary arteries, or both, as determined by extensive dissection.

Anesthesia was induced with ketamine (5.0 mg/kg, IV) and diazepam (0.25 mg/kg, IV) and maintained with 1.0% to 2.5% halothane in oxygen. Following endotracheal intubation, the dogs were mechanically ventilated. Fluid-filled catheters were placed into the femoral artery and vein. The femoral venous catheter allowed administration of fluids or drugs as needed.

Following median sternotomy, a hemispheric hollow capsule7,9 was attached to the epicardial surface of the left ventricle near the apex by use of cyanoacrylate adhesive (Figure 1). The capsule enclosed an epicardial surface of 0.636 cm2. Care was taken to avoid placing the capsule over areas with visible subepicardial vessels. Two ports extended from the capsule chamber. One connected via polyethylene tubing (1.14 mm, ID) to a glass pipette graduated in 0.01 mL and the other connected the chamber through polyethylene 160 tubing and a 3-way stopcock to a pressure transducer.a The distal end of the glass pipette was open to room air. Once the capsule was adhered to the epicardium, the system was filled with a solution of bovine albuminb in saline (0.9% NaCl) solution. Movement of the fluid meniscus within the pipette was measured and recorded as an indicator of movement of fluid across the epicardium into or out of the capsule. This system allowed capsule pressure and epicardial flow to be measured simultaneously. Capsule pressure was manipulated by adjusting the height of the glass pipette.

Figure 1—
Figure 1—

Diagram of an apparatus for determination of transepicardial flow in dogs. A hemispheric plastic capsule is attached to the left ventricular epicardial surface and filled with an albumin-saline (0.9% NaCl) solution. Fluid-filled tubing extending from the capsule allows measurement of intracapsular hydrostatic pressure (PP), which represents pericardial hydrostatic pressure, and flow into or out of the capsule (QP), which represents transepicardial flow (JV,e).

Citation: American Journal of Veterinary Research 69, 3; 10.2460/ajvr.69.3.356

Physiologic measurements—Pressure in the femoral artery was measured via fluid-filled catheters connected to pressure transducers.a Femoral artery pressure was used as one of the indicators of depth of anesthesia. Data from the epicardial capsule pressure transducer were recorded through a transducer amplifier and analogue-digital converterc directly to a computerd running data acquisition software.e

Protein concentrations of the pericardial fluid samples were determined by refractometry.f To facilitate measurement of protein concentrations < 2.5 g/dL, total solids concentration was plotted as a function of refractive index by use of data from the scales of the refractometer. This relationship is linear, which permitted measurement of the refractive index of each sample and calculation of its protein concentration by use of the derived relationship.

Determination of the COP of the intracapsular albumin-saline solutions was simplified by construction of a reference curve relating refractive index of the solution to its COP. An albumin-saline solution was serially diluted with saline solution from a starting albumin concentration of approximately 5 g/dL to a final concentration of approximately 0.6 g/dL. The refractive index and COP were determined for each dilution. The COP of the intracapsular fluid was determined with a colloid osmometer.g This process was repeated 4 times. The relationship between refractive index and COP was determined by plotting COP as a function of refractive index and determining a best-fit regression. Subsequently, the COP for any albumin-saline solution used in an experiment was estimated by determining the solution's refractive index and calculating the COP according to the derived relationship.

Myocardial hydroxyproline content was assessed as an index of collagen content of the left and right ventricular free walls. Hydroxyproline content was determined by homogenizing a small (approx 10 mg) sample of dried myocardium with a small, known volume of distilled water. A portion of this homogenate (4 to 5 mg) was accurately weighed and transferred into a 13 × 100-mm screw-cap tube. Two milliliters of 6N HCl was pipetted into the tube. This mixture was incubated at 110°C for 16 hours. The mixture was titrated to a pH between 4 and 6 with up to 2 mL of 6N NaOH. Two reagent solutions were prepared. Reagent A was made from chloramine T (7% wt/vol in distilled water) and an acetate-citrate buffer (57.0 g of anhydrous sodium acetate, 37.5 g of trisodium citrate, 5.5 g of citric acid, and 385 mL of isopropanol in a final volume of 1,000 mL [with distilled water]) in a solution of 1:4 (vol/vol) chloramine T:acetate-citrate buffer. Reagent B was a 3:13 (vol/vol) solution of dimethylaminobenzaldehyde: isopropanol. The dimethylaminobenzaldehyde (Ehrlich reagent) solution consisted of 20 g of dimethylaminobenzaldehyde in 30 mL of 60% perchloric acid.

Four hydroxyproline standards (1, 5, 10, and 25 μg/mL) were then processed in a fashion identical to the samples. A 200-μL aliquot of each standard or sample was pipetted into 10 × 100-mm tubes. A 400-μL aliquot of isopropanol was added and vortexed. A 200-μL aliquot of reagent A was added, and the mixture was incubated at 25°C for 4 minutes. A 2.5-mL aliquot of reagent B was then added, and the solution was mixed with a vortex mixer. The tube was covered and incubated at 58°C for 25 minutes. The tube was cooled under running tap water for 3 minutes and assessed on a spectrophotometer at 558 nm within 30 minutes. A tube of distilled water was used as a blank for spectrophotometry.

A standard curve was generated by plotting the absorbance of the standard solution as a function of its hydroxyproline concentration. The hydroxyproline concentration of the samples was then calculated from the measured absorbance with the standard curve. The hydroxyproline concentration was divided by the volume added to the dried sample and by the weight of the dried sample to provide the hydroxyproline content of each sample as a ratio of total dry matter (mg/g).

Protocol—The JV,e was measured at each of 3 capsule pressures (−5, −15, and −25 mm Hg) and with 2 solutions of bovine albumin in saline solution in the capsule with albumin concentrations of 2.2 (solution 1; COP = 7.0 mm Hg) and 0.6 g/dL (solution 2; COP = 2.0 mm Hg). Each JV,e measurement was made over a 30-minute period. The measurement period was begun approximately 10 minutes after the capsule pressure was established to allow a period of equilibration. Between each measurement period, fresh solution was gently flushed through the system. The order of the pressure settings and the intracapsular solutions used was randomized.

Following each experiment, the epicardial capsule system was carefully examined for leaks and to ensure that the adhesive did not cover any portion of the epicardial surface within the capsule. Data from experiments in which these defects were discovered were not used.

Following completion of the JV,e measurements, each dog was euthanized while anesthetized with a bolus of concentrated potassium chloride solution administered IV. The heart was removed, and the right ventricle and pulmonary arteries were examined for adult heartworms. The right and left ventricular free walls were removed by transecting along the junction of the septum and free wall and along the coronary groove. The ventricular free walls were weighed and dried to a constant weight in a drying oven at 60°C. The dry weight of each specimen was recorded, and the myocardial water content for each specimen was calculated as (wet weight − dry weight)/dry weight, which yields the water-to-dry weight ratio.

Statistical analysis—All data are presented as mean ± SEM. Transepicardial flux (JV,e/Ae) was calculated as JV,e (μL/h) divided by the surface area of epicardium within the capsule (0.636 cm2). The JV,e/Ae was plotted as a function of capsule pressure for each capsule solution and for each group (heartworm-free and heartworm infected), analyzed by use of linear regression, and compared by use of 2-factor ANCOVA. Myocardial water-to-dry weight ratio and hydroxyproline content data were compared by use of unpaired t tests.

The ANCOVA analyzes the effect of a factor on 2 groups of data, each describing the relationship between an independent variable (covariate or regressor) and a dependent variable (response).10 In this experiment, capsule pressure was the covariate, epicardial flux was the response, and capsule solution was the factor. First, the assumption that the slopes of the relationships are not different is tested.10 If the slopes are significantly different, no further analysis is performed. If no difference is detected, the assumption that the response means of the 2 groups are the same is tested. Comparison of the response means requires that the means be adjusted for the effect of the covariate on the response. The difference between the adjusted means is termed the adjusted mean difference. This difference is analogous to the distance along the y-axis between 2 parallel lines. A value of P < 0.05 was considered significant.

Results

Data were summarized (Table 1). Two to 4 adult heartworms were found in the right ventricle of 7 of the dogs (4 females and 3 sexually intact males). The remaining 8 dogs (4 females and 4 sexually intact males) were free of adult heartworms. Pericardial fluid protein concentration in heartworm-free dogs (1.49 ± 0.23 g/dL) was not significantly different from that in heartworm-infected dogs (1.26 ± 0.14 g/dL).

Table 1—

Summary data for myocardial and epicardial fluid balance in 8 heartworm-free and 7 heart-worm-infected dogs.

VariableHeartworm-freeHeartworm-infected
Body weight (kg)20.1 ± 1.021.1 ± 1.4
Pericardial fluid protein concentration (g/dL)1.49 ± 0.231.26 ± 0.14
Epicardial hydraulic conductance - Soln 1 (μL/h/cm2/mm Hg)3.6 ± 0.63.3 ± 0.5
Epicardial hydraulic conductance - Soln 2 (μL/h/cm2/mm Hg)3.8 ± 0.83.4 ± 0.5
LV water-to-dry weight ratio3.32 ± 0.023.23 ± 0.02*
RV water-to-dry weight ratio3.17 ± 0.042.96 ± 0.06*
LV hydroxyproline content (mg/g)3.10 ± 0.213.14 ± 0.23
RV hydroxyproline content (mg/g)4.57 ± 0.304.63 ± 0.39

Significantly (P < 0.05) different from value in heartworm-free dogs.

LV = Left ventricle. RV = Right ventricle. Soln 1 = Solution 1. Soln 2 = Solution 2.

The epicardial capsule in 1 preparation developed a leak during the experiment; however, the data collected prior to that point were not compromised and were included in the results. Replacement of the capsule without invalidating the data is difficult because the cyanoacrylate adhesive on the epicardial surface from the original placement impedes transepicardial flow. The intracapsular hydrostatic pressure in all preparations cycled in concert with the heartbeat. Mean intracapsular hydrostatic pressure for each 30-minute measurement period was used to represent Pp for calculation of hydraulic conductance (Lp,e) in equation 2.

Transepicardial flux (JV,e) was plotted as a function of Pp for solutions 1 and 2 for both groups of dogs (Figure 2). A significant negative correlation was detected between Pp and JV,e. The slopes of the regression lines through the 4 data sets were not significantly different and equaled −3.6 ± 0.6 μL/h/cm2/mm Hg and −3.8 ± 0.8 μL/h/cm2/mm Hg for solutions 1 and 2, respectively, in the heartworm-free group and −3.3 ± 0.5 μL/h/cm2/mm Hg and −3.4 ± 0.5 μL/h/cm2/mm Hg for solutions 1 and 2, respectively, in the heartworm-infected group. Therefore, no difference was detected in epicardial water permeability (LP,e) on the basis of heartworm status.

Figure 2—
Figure 2—

Effect of hydrostatic pressure in the epicardial capsule on transepicardial flow of capsule solutions 1 (closed circles and solid line; COP = 7.0 mm Hg) and 2 (open circles and dashed line; COP = 2.0 mm Hg) in 8 heartworm-free (A) and 7 heartworm-infected (B) dogs. (Data in part A adapted from Stewart RH, Rohn DA, Allen SJ, et al. Basic determinants of epicardial transudation. Am J Physiol Heart Circ Physiol 1997;273:H1408–H1414. Reprinted with permission from the American Physiological Society.)

Citation: American Journal of Veterinary Research 69, 3; 10.2460/ajvr.69.3.356

The JV,e for solution 1 was significantly greater than that for solution 2 (adjusted mean difference for pooled data, 15.3 ± 4.8 μL/h/cm2). However, this effect of solution on flux was not significantly different between heartworm-infected dogs (adjusted mean difference, 14.2 ± 5.8 μL/h/cm2) and heartworm-free dogs (16.5 ± 8.4 μL/h/cm2). These findings indicated that transepicardial flux was significantly affected by the albumin concentration and, thus, the COP (0p) of the 2 capsule solutions, as predicted by equation 1. However, no difference in the permeability of the epicardium to protein (ie, colloid) was detected on the basis of heartworm status. The reflection coefficient (σd,e) calculated from equation 3, in which b1-2 was the adjusted mean difference and 0p,1 − 0p,2 was 5.0 mm Hg, equaled 0.90 for heartworm-free dogs and 0.86 for heartworm-infected dogs. Driving pressure was plotted as a function of myocardial water-to-dry weight ratio (Figure 3).

Figure 3—
Figure 3—

The effect of heartworm status on interstitial driving pressure and myocardial water content. Interstitial driving pressure (Pint − σd,e0int) is plotted as a function of myocardial water-todry weight ratio for heartworm-infected dogs (black circle) and heartworm-free dogs (white circle). A regression line (dashed line) derived from Laine and Granger14 has been added to indicate the expected relationship between myocardial interstitial hydrostatic pressure and water content. Data are expressed as mean ± SEM.

Citation: American Journal of Veterinary Research 69, 3; 10.2460/ajvr.69.3.356

Left ventricular myocardial water-to-dry weight ratio was significantly greater in the heartworm-free dogs (3.32 ± 0.02) than in the heartworm-infected dogs (3.23 ± 0.02). The water-to-dry weight ratio of the right ventricular free wall was also significantly greater in the heartworm-free dogs (3.17 ± 0.04 vs 2.96 ± 0.06). No significant differences between the groups were found in myocardial hydroxyproline-to-dry matter ratio of the free walls of the left or right ventricle (Table 1).

Discussion

The data indicated that not only was the myocardium of the heartworm-infected dogs not edematous but the myocardial water content was significantly lower in heartworm-infected dogs than in heartwormfree dogs. No differences between the groups were detected regarding the myocardial hydroxyproline (collagen) content, pericardial fluid protein concentration, or permeability of the epicardium to water or protein, as indicated by the coefficients in the Starling-Landis equation, hydraulic conductance (Lp,e), and the reflection coefficient (σd,e).

The myocardial water-to-dry weight ratios in the present study were consistent with those reported by Laine and Allen1 (left ventricle, 2.90 ± 0.20), Pratt et al4 (left ventricle, 3.52 ± 0.05; right ventricle, 3.52 ± 0.01), Davis et al2 (left ventricle, 3.41 ± 0.17), and Mehlhorn et al11 (left ventricle, 3.15 ± 0.14). In addition, the value for left ventricle myocardial hydroxyproline-todry matter ratio was similar to that reported by Laine and Allen (5.23 ± 0.84 mg/g [reported as a collagen-todry matter ratio of 3.90 ± 0.64 g/100 g]). The values of pericardial fluid protein concentration measured in the present study were also similar to previous reports of 1.65 ± 0.07 g/dL measured by use of colorimetric methods12 and 1.70 g/dL measured by use of refractometry.13 The albumin concentrations of the 2 epicardial capsule solutions used in the study, 0.6 and 2.2 g/dL, were similar to the lowest and highest pericardial fluid protein concentration values observed in the heartworm-free dogs (0.89 and 2.35 g/dL) and therefore represented physiologic values.

The flow of myocardial interstitial fluid across the epicardium is sensitive to acute changes in myocardial water content.7 Increased coronary sinus pressure both induces myocardial interstitial edema formation and significantly increases transepicardial fluid flow.1,7 This increase in transepicardial flow acts as a protective mechanism against acute myocardial edema formation. In the present study, however, despite a difference in myocardial water content, no difference in transepicardial flow could be detected between the groups. This suggests that not only was epicardial permeability to water and protein the same in both groups but also interstitial driving pressure for that flow (Pint − σd,eπint) was not different.

To explore this possibility, the interstitial driving pressure was derived by rearranging equation 1 to reveal that when JV,e is equal to zero, then the following equation is used:

article image

The relationship between transepicardial flow and capsule pressure was analyzed for each dog and each capsule solution by use of linear regression, and the capsule pressure (Pp) that corresponded to zero flow was determined. That value, the COP of the capsule solution, and the refection coefficient were used to calculate driving pressure. Those estimates for driving pressure, which is the effective difference between hydrostatic pressure and COP within the interstitial space, ranged from −20 to −4 mm Hg. Driving pressure was plotted as a function of myocardial water-to-dry weight ratio. Results indicated that although the myocardial tissue of heartworm-infected dogs had lower water content than heartworm-free dogs, the interstitial driving pressure appeared to have been greater. A line that indicated the expected relationship between myocardial water content and interstitial hydrostatic pressure was added to the plot.14

The data from the heartworm-infected dogs appeared to represent a change in the interstitial pressure-volume relationship from that of heartworm-free dogs. This change could be explained by the findings of recent research indicating that the interstitial pressurevolume relationship is actively regulated through alterations in the structure of the extracellular matrix.15,16 This structural alteration can occur acutely and does not require deposition of interstitial collagen. The extracellular matrix in many tissues exists in a relatively dehydrated state, in part, because fibroblasts within the matrix individually exert a traction force on multiple collagen fibers, thus compressing the matrix.17,18 When this traction is disrupted by inflammatory mechanisms, the matrix expands and the interstitial water content increases without a concomitant increase in interstitial fluid pressure.15 This type of structural change that results in alteration of the interstitial pressure-volume relationship may also explain a lower interstitial water volume with relatively normal interstitial pressure; however, little information is available regarding the effects of chronic tissue remodeling on the interstitial pressure-volume relationship.

The lower myocardial water content in heartworminfected dogs could be explained by an increase in myocardial dry weight as well as a decrease in myocardial water weight. An increase in dry weight could occur if the presence of heartworms was associated with development of myocardial interstitial fibrosis. In a study of left ventricular myocardial edema, Laine and Allen1 detected interstitial fibrosis characterized by a significant increase in myocardial hydroxyproline content in dogs with increased right heart elevation of 2 months' duration caused by pulmonary artery banding. However, by use of myocardial hydroxyproline content as an index of collagen content in the present study, evidence for interstitial fibrosis in either the right or left ventricular free walls in heartworm-infected dogs was not detected.

Data of the present study are not consistent with results of a previous investigation of naturally acquired heartworm infection in dogs that underwent cardiopulmonary bypass.6 In that study, heartworminfected dogs had significantly greater myocardial water-to-dry weight ratio (3.50 ± 0.07), compared with control dogs (2.95 ± 0.09).6 Extravascular fluid content was determined in a fashion similar to the present study. One explanation for this difference is that in the previous study,6 the measured ventricular wet weights for blood volume were adjusted by use of a spectrophotometric correction. However, Laine and Allen1 were unable to detect a significant difference in blood as a percentage of left ventricular wet weight between normotensive dogs and dogs with myocardial edema associated with a chronic increase in right heart pressure. It is, therefore, unlikely that the blood-free correction could have such a severe effect on estimates of extravascular fluid content. A second reason for the difference could be the small number of animals (2 to 3/group) used in the previous study.6 A third reason could be that both the present study and the previous study included dogs with naturally acquired infections in which duration and severity of the infections were not controlled.

The heartworm-infected dogs were grouped as such on the basis of adult heartworms found on necropsy rather than on the basis of clinical signs. Physical examination of the dogs revealed no evidence of disease; however, detailed antemortem examination of the cardiovascular system including determination of cardiac and vascular pressures or functional properties was not performed. Therefore, the degree of cardiovascular disease in the infected dogs was unknown.

The data reported here suggest that heartworm infection was associated with a change in the interstitial pressure-volume relationship that resulted in lower interstitial fluid volume at a similar interstitial fluid pressure. This change did not appear to affect cardiac function to a degree sufficient to induce overt cardiovascular disease. However, it appeared to render the dogs more vulnerable to other stressors of the cardiovascular system as evidenced by the enhanced development of myocardial edema and poor clinical outcome of infected dogs placed on cardiopulmonary bypass.6 Therefore, evidence of heartworm infection should be considered a risk factor in dogs, independent of overt evidence of cardiovascular dysfunction.

ABBREVIATIONS

COP

Colloid osmotic pressure

a.

Isotec, Healthdyne Cardiovascular, Irvine, Calif.

b.

Fraction V, 96-99% bovine albumin, Sigma Chemical Co, St Louis, Mo.

c.

MacLab, Division of ADInstruments, Milford, Mass.

d.

Macintosh Quadra 950, Apple Computer Inc, Cupertino, Calif.

e.

Chart, MacLab, Division of ADInstruments, Milford, Mass.

f.

AO TS Meter, American Optical, Buffalo, NY.

g.

Model 4400, Wescor Inc, Logan, Utah.

References

  • 1.

    Laine GA, Allen SJ. Left ventricular myocardial edema: lymph flow, interstitial fibrosis, and cardiac function. Circ Res 1991;68:17131721.

  • 2.

    Davis KL, Mehlhorn U, Laine GA, et al. Myocardial edema, left ventricular function, and pulmonary hypertension. J Appl Physiol 1995;78:132137.

  • 3.

    Mehlhorn U, Davis KL, Burke EJ, et al. Impact of cardiopulmonary bypass and cardioplegic arrest on myocardial lymphatic function. Am J Physiol 1995;268:H178–H183.

    • Search Google Scholar
    • Export Citation
  • 4.

    Pratt JW, Schertel ER, Schaefer SL, et al. Acute transient coronary sinus hypertension impairs left ventricular function and induces myocardial edema. Am J Physiol 1996;271:H834–H841.

    • Search Google Scholar
    • Export Citation
  • 5.

    Miyamoto M, McClure DE, Schertel ER, et al. Effects of hypoproteinemia-induced myocardial edema on left ventricular function. Am J Physiol 1998;274:H937–H944.

    • Search Google Scholar
    • Export Citation
  • 6.

    Rohn DA, Davis KL, Mehlhorn U, et al. Myocardial edema and compromised left ventricular function attributable to dirofilariasis and cardiopulmonary bypass in dogs. Am J Vet Res 1995;56:221226.

    • Search Google Scholar
    • Export Citation
  • 7.

    Stewart RH, Rohn DA, Allen SJ, et al. Basic determinants of epicardial transudation. Am J Physiol Heart Circ Physiol 1997;273: H1408–H1414.

    • Search Google Scholar
    • Export Citation
  • 8.

    Michel CC. Fluid movements through capillary walls. In: Renkin EM, Michel CC, eds. Handbook of physiology, section 2: the cardiovascular system. Vol IV. New York: Oxford University Press, 1984;375409.

    • Search Google Scholar
    • Export Citation
  • 9.

    Negrini D, Townsley MI, Taylor AE. Hydraulic conductivity of canine parietal pleura in vivo. J Appl Physiol 1990;69:438442.

  • 10.

    Mason RL, Gunst RF, Hess JL. Analysis of covariance for completely randomized designs. In: Statistical design and analysis of experiments: with applications to engineering and science. New York: Wiley, 1989;398406.

    • Search Google Scholar
    • Export Citation
  • 11.

    Mehlhorn U, Allen SJ, Adams D, et al. Normothermic continuous antegrade blood cardioplegia does not prevent myocardial edema and cardiac dysfunction. Circulation 1995;92:19391946.

    • Search Google Scholar
    • Export Citation
  • 12.

    Gibson AT, Segal MB. A study of the composition of pericardial fluid, with special reference to the probable mechanism of fluid formation. J Physiol 1978;277:367377.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13.

    Maurer FW, Warren MF, Drinker CK. The composition of mammalian pericardial and peritoneal fluids. Am J Physiol 1940;128:635644.

  • 14.

    Laine GA, Granger HJ. Microvascular, interstitial and lymphatic interactions in normal heart. Am J Physiol 1985;249:H834–H842.

  • 15.

    Reed RK, Rodt SA. Increased negativity of interstitial fluid pressure during the onset stage on inflammatory edema in rat skin. Am J Physiol 1991;260:H1985–H1991.

    • Search Google Scholar
    • Export Citation
  • 16.

    Wiig H, Rubin K, Reed RK. New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences. Acta Anaesthesiol Scand 2003;47:111121.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17.

    Grinnell F, Lamke CR. Reorganization of hydrated collagen lattices by human skin fibroblasts. J Cell Sci 1984;66:5163.

  • 18.

    Berg A, Ekwall A-KH, Rubin K, et al. Effect of PGE1, PGI2, and PGF2 A analogs on collagen gel compaction in vitro and interstitial pressure in vivo. Am J Physiol 1998;274:H663–H671.

    • Search Google Scholar
    • Export Citation

Contributor Notes

Supported by National Heart, Lung, and Blood Institute grants HL-36115, HL-01999, and HL-077566 and the American Heart Association.

Address correspondence to Dr. Stewart.