INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

AN INCREASED WORKLOAD imposed on the heart by a sustained volume overload induces compensatory cardiomyocyte hypertrophy and ventricular dilatation (12, 31). In a previous study, we described an early and progressive decrease in myocardial contractility, marked hypertrophy, and ventricular dilatation in response to a biventricular volume overload induced by infrarenal aorto-caval (AV) fistula (5). There was also an initial increase in in vivo left ventricular (LV) end-diastolic pressure (EDP), which peaked at 3 wk. A subsequent decline in in vivo LVEDP was associated with significant ventricular dilatation and increased LV compliance. Approximately 50% of the rats at the 8-wk postfistula time point developed symptoms consistent with congestive heart failure (CHF). However, it was not clear from that study whether the compensatory ventricular remodeling induced by an AV fistula would continue beyond 8 wk or whether the remaining animals would have subsequently developed overt CHF. Therefore, the purpose of this study was to determine the temporal response of ventricular remodeling and function beyond 8 wk postfistula and the morbidity and mortality due to CHF in this model of sustained volume overload. To this end, LV mass, dilatation, compliance, function, matrix metalloproteinase (MMP) activity, and collagen concentration of hearts from rats with chronic AV fistula for 15 and 21 wk and controls were compared. The findings of this study provide experimental verification of the hypothesis first put forward by Linzbach (21) that the transition to heart failure is triggered by marked ventricular dilatation occurring once the myocardial hypertrophic response is exhausted. Together these studies characterize the temporal progression of ventricular hypertrophy and size and morbidity and mortality, and they clearly establish the AV fistula in rats as an appropriate model to study the pathogenesis of CHF.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

All experiments were performed using adult male Sprague-Dawley (Hsd:SD) rats initially weighing between 380 and 440 g. The rats were housed under standard environmental conditions and maintained on commercial rat chow and tap water ad libitum. All studies conformed with the principles of the National Institutes of Health Guide for the Care and Use of Laboratory Animals, and the protocol was approved by our Institution's Animal Care and Use Committee. Anesthesia for surgical procedures and subsequent euthanasia at the experimental end point was effected by pentobarbital sodium (50 mg/kg) injected into the intraperitoneal cavity. Postoperative analgesia was provided by buprenorphine HCl (0.05 mg/kg sc) administered to the rats at the time of surgery.

Surgical preparation. Infrarenal abdominal AV fistula was created in rats as previously described (5). Briefly, a ventral abdominal laparotomy was performed to expose the aorta and caudal vena cava ~1.5 cm below the renal arteries. An 18-gauge needle was inserted into the exposed abdominal aorta and advanced through the medial wall into the vena cava to create the fistula. The needle was withdrawn, and the ventral aortic puncture was sealed with cyanoacrylate. Creation of a successful AV fistula was visually evident by the pulsatile flow of oxygenated blood into the vena cava. The abdominal musculature and skin incisions were closed by standard techniques with absorbable suture and autoclips, respectively.

Experimental protocol. Rats were randomly selected from the colony and studied at 15 (n = 7 rats) and 21 wk (n = 9 rats) following creation of the AV fistula. Age-matched control groups for each study period consisted of sham-operated rats (n = 7 each time period). At the end of the study period, the rat was weighed and anesthetized, the fistula patency was visually confirmed, and the heart was removed and attached to a perfusion apparatus for evaluation of ventricular function (see Assessment of ventricular size and function). After the functional studies were completed, the atria and great vessels were removed and the LV (plus septum) and right ventricle (RV) were separated and weighed. Lung wet weight was also obtained after the esophagus and trachea were trimmed away, and the pleural surface was blotted dry. Results from an 8-wk postfistula (n = 16) and corresponding control group (n = 7) that were previously published (5) are included herein for comparison purposes.

Assessment of ventricular size and function. LV volume and function were evaluated in vitro using a blood-perfused isolated heart preparation as previously described (5). Briefly, the apparatus consisted of a pressurized perfusion reservoir and a collection reservoir connected in circuit with a support rat. Arterial blood from the carotid artery of the support rat was pumped to a pressurized reservoir for retrograde perfusion of the heart. The coronary venous effluent was collected and returned to the support rat through a jugular vein catheter to filter and oxygenate the blood supply to the isolated heart. The temperature of the blood in the perfusion reservoir was maintained at 37 ± 1°C, and the environment around the isolated heart was kept constant at 35 ± 2°C. Coronary perfusion pressure was maintained between 100 and 110 mmHg.

Morbidity and mortality assessment. To determine the time course of morbidity and mortality in this model, 116 rats with AV fistula were evaluated weekly for signs of CHF. For the purpose of this study, overt CHF (i.e., morbidity) was defined as a significant increase in body weight (i.e., 50 g in a 7-day period) together with labored respiration and/or pitting edema. Once symptoms of overt CHF were identified, LV function was evaluated as described above. The temporal pattern of survival postfistula was plotted as the percentage of rats remaining alive that had not developed CHF or succumbed to sudden death (i.e., dividing the number of rats alive each week postfistula by 116).

Histological evaluation. After the ventricles were weighed, a complete cross section was taken from the midportion of each ventricle and placed in buffered formalin for fixation. The tissue was then processed for routine histopathology, and 5-µm-thick paraffin-embedded sections were stained with hematoxylin and eosin for evaluation of myocardial morphology and evidence of necrosis and picrosirius red F3BA for evaluation of the myocardial interstitial collagen. LV interstitial collagen volume fraction (CVF) was determined from the picrosirius red F3BA-stained sections by using a Quantimet 520 Image Analysis System (Leica; Deerfield, IL) and established methods (40). Perivascular collagen and scars were excluded from the analysis of CVF.

MMP activity. MMP activity in cardiac tissue extracts was determined by gelatin zymography performed by standard procedures by using a sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) matrix containing gelatin (1 mg/ml) (38). Cardiac tissue extracts were prepared by washing 40-50 mg of frozen LV in cold saline. The tissue was then minced into 1-mm fragments, which were agitated for 48 h at 4°C in 1 ml of ice-cold extraction buffer containing 10 mmol/l cacodylic acid, 150 mmol/l NaCl, 1 µmol/l ZnCl₂, 20 mmol/l CaCl₂, 3.0 mmol/l NaN₃, and 0.01% Triton X-100. Upon completion of the extraction incubation, the supernatant was decanted and added to 200 µl of 0.1 mol/l Tris buffer, and the samples were centrifuged (4°C, 15 min, 13,000 g). The final protein concentration of the tissue extract was determined with a standardized colorimetric assay (Bio-Rad Protein Assay), and the extracted samples were then aliquoted and stored at 70°C. Each sample was mixed with Laemmli SDS sample buffer in the absence of reducing agent and electrophoresed on a 10% polyacrylamide gel at 20°C. After electrophoresis, the gels were washed four times for 20 min in 2.5% Triton X-100 to remove the SDS and permit enzyme renaturation. The gel was then placed in substrate buffer [40 mM Tris · HCl, 200 mM NaCl, 5 mM CaCl₂, 0.02% (wt/vol) Brij-35, and 0.02% sodium azide, pH 7.4] and incubated on a shaker for 18 h at 37°C. The gels were then stained with 0.1% Coomassie Brilliant Blue and destained in H₂O, and activity of the MMP bands was quantified by densitometry (ImageQuant, Molecular Dynamics). All of the zymograms had two lytic bands corresponding to standards for gelatinase A (MMP-2) proenzyme (68 kDa) and activated (62 kDa) forms (Chemicon International; Temecula, CA). Less consistent was the appearance of a lytic band at 92 kDa consistent with gelatinase B (MMP-9).

Data and statistical analysis. LV volumes were adjusted to account for the volume displaced by the balloon material in the LV lumen; balloon volume was calculated from the weight of the balloon divided by the density of the latex (0.898 g/ml). Pressure-volume curves generated for the balloons alone indicated that their contribution to the LVEDP measurement was negligible (i.e., <1 mmHg) for the volume range 0-1,240 µl. Statistical analyses were performed using SYSTAT 7.0 software (SPSS: Chicago, IL). All grouped data are expressed as means ± SD. Grouped data comparisons were made by one-way analysis of variance. When a significant F ratio (P < 0.05) was obtained, intergroup comparisons were made using a modified t-test and Bonferroni bounds. Statistical significance was taken to be P < 0.05/k, where k equaled the number of comparisons. Before the peak isovolumetric pressure-volume relationships for each study group were compared, the pressure-volume data for individual hearts were adjusted so that all pressure-volume curves began at the origin of the graph. The adjustment was made by subtracting V₀ and peak isovolumetric pressure at V₀ (P₀), respectively, from all subsequent volume and pressure data.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Control groups. No significant trends were observed in LV, RV, and lung weights in the individual control groups (Table 1). Body weight increased with age, but these differences were not statistically significant. Statistical comparisons of the EDP-volume and peak isovolumetric pressure-volume relationships from the individual age-matched control groups (data not shown) revealed no significant differences. Therefore, these data were combined and considered as a single group (control) for comparison purposes.

AV fistula groups. Average LV, RV, lung, and body weights for the control and volume overload groups are presented in Table 1. There was a significant increase in both LV and RV weights (P < 0.001) relative to the age-matched controls. Most of the myocardial hypertrophy induced by the AV fistula occurred during the first 8 wk postfistula, but there was a trend for ventricular weights to increase further with time, although this was only significant for the RV weights of the 15- and 21-wk groups relative to the 8-wk AV fistula group. The average lung weight in all of the AV fistula groups was significantly increased over control (P  0.05), and there was a tendency for the relative increase in lung weights to be much larger beyond 8 wk postfistula. Significant increases in body weight above the age-matched control groups were identified in the 8- and 15-wk groups, but at 21 wk postfistula the body weight was comparable to control. Because the heart weight-to-body weight ratio is of limited value as an index of hypertrophy in the presence of systemic edema, it was not calculated.

LV diastolic function. The in vitro LV volumes over the EDP range of 0-25 mmHg are presented in Table 2, and the average LV EDP-end-diastolic volume (EDV) curves for the control and AV fistula groups are shown in Fig. 1. The LV EDP-EDV relationship was shifted significantly to the right at all three times postfistula relative to the control relationship. This shift in the pressure-volume relationship was due to both ventricular dilatation and an overall increase in ventricular compliance. An assessment of ventricular dilatation was made by comparing the values for V₀, whereas an indication of ventricular compliance was ascertained from the volume required to increase EDP from 0 to 25 mmHg (V_0-25). Thus it can be seen from the data in Table 2 and Fig. 1 that, although the LV in the AV fistula groups are significantly dilated and much more compliant than the normal hearts, there was no further shift to the right in the LV pressure-volume relationships beyond 8 wk postfistula (when compared strictly on a time postfistula basis).

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Temporal response of the left ventricular (LV) end-diastolic pressure-volume relationships to chronic volume overload. The pressure-volume relationships for hearts from the age-matched control groups are presented as a single average combined group. The control group is significantly different from the 8, 15, and 21 wk aorto-caval (AV) fistula groups (P < 0.01).

LV systolic function. Measures of LV systolic function, namely P₀ and peak isovolumetric pressure at an LVEDP of 25 mmHg (P₂₅); the peak isovolumetric pressure-EDV (P_max-V) relationship slope, which is an index of contractility; and the range of regression coefficients for the P_max-V relationship are given in Table 3 for each group. There was a tendency for the P₀ and P₂₅ values to be slightly lower in the AV fistula groups relative to control, but this decrease did not attain statistical significance. The relationship between peak isovolumetric pressure and EDV was highly linear, as evidenced by correlation coefficients, which were all >0.88 with the exception of one curve in the 8-wk group. The slopes of the normalized P_max-V relationships for all of the fistula groups were significantly less than control (P < 0.05), indicating a decrease in contractility.

Morbidity and mortality. The incidence of symptoms or death attributable to CHF can be derived from the plot of survival postfistula depicted in Fig. 2. The initial occurrence of morbidity and mortality due to CHF was noted at 4 wk after inducing chronic volume overload, with the incidence of overt CHF subsequently exceeding 50% and 80% at 14 and 21 wk postfistula, respectively. Thirty-three of the rats (28%) died acutely without noticeable increases in body weight; however, almost all of these rats had lungs that did not collapse when the thorax was opened, together with marked increases in lung weight and lungs with a wet, heavy appearance; typically, there was also excess fluid in the thoracic cavity. No adverse events occurred in the control group during the same period.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Percentage of 116 rats with an AV fistula surviving each week. The decline in percent survivors reflects the progressive development of congestive heart failure (CHF) and the associated morbidity and mortality induced by chronic volume overload.

Compensated volume overload versus overt heart failure. These experiments were initially designed to progress to a specific temporal end point; however, comparisons based solely on time postfistula revealed no identifiable trends with regard to the development of overt CHF. Nevertheless, it was clear that individual rats were developing CHF after differing periods of compensated LV remodeling. Therefore, each rat from the 8-, 15-, and 21-wk fistula groups was separated into compensated or failure groups based on the presence or absence of symptomatic CHF. The average values obtained for the morphological and functional parameters based on this grouping are presented in Fig. 3 and Tables 1-3. In comparing the compensated and failure groups, there was, as expected, a marked increase in lung weight observed in rats with overt CHF. The hypertrophic response in the RV was also correlated with lung weight (r = 0.76). The initial RV hypertrophy after opening the fistula stabilized after an ~100% increase in mass. This initial rate of RV hypertrophy roughly paralleled that of the LV and was typically associated with relatively normal lung weights. However, coincident with the development of LV dysfunction, as evidenced by a marked increase in lung weight, the extent of RV hypertrophy became disproportionately greater than that of the LV, as reflected by a significant decrease in the LV-to-RV ratio (3.20 ± 0.20 to 2.69 ± 0.33, P < 0.001). The duration of time postfistula before rats developed symptoms of overt CHF was correlated with LV weight (r = 0.75). However, surprisingly there was very little difference between the compensated and failing hearts with regard to LV weight (P = 0.92), with the LV weights being comparable in both the compensated and overt CHF groups, ranging from 1,252 to 2,279 mg and 1,425 to 2,009 mg, respectively.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   LV end-diastolic pressure-volume (P-V) relationships grouped according to the presence or absence of CHF. Dashed line is the P-V curve for the 21-wk group in Fig. 1. The hearts from rats having no symptoms of heart failure have relatively modest ventricular dilatation and increased compliance compared with the hearts from rats in overt heart failure.

Histological evaluation. As can clearly be seen from Fig. 4, the fistula hearts were grossly larger, but other than an obvious cardiomyocyte hypertrophy, differences from normal hearts at the light microscopic level were subtle. Muscle fiber thinning and widened interstitial spaces were seen in failing hearts, with the latter being attributable to the diffuse systemic edema characteristic of CHF. There was no evidence of myocyte necrosis, inflammatory infiltrates, or scar tissue deposition, and the occasional foci of mild to moderate interstitial fibrosis localized to the subendocardium and papillary muscles present in a few of the hearts was consistent with the pattern of cardiac fibrosis described as an incidental aging lesion in rats (2). However, there was a generalized increase in the myocardial interstitial collagen concentration in the rats with overt CHF (CVF of 1.42 ± 0.26% compared with a control value of 1.02 ± 0.39%, P < 0.01), whereas the compensated rats had CVF values comparable to control (1.08 ± 0.35%). There were no other consistent differences in the histological appearance of the myocardium among the normal, compensated, and overt heart failure groups.

View larger version (105K): [in this window] [in a new window]   Fig. 4.   Remodeling secondary to volume overload is characterized by ventricular dilatation and wall thinning. Photo depicts representative hearts from a control rat (left) and a rat 16 wk postfistula (right) that were perfusion fixed at in vivo mean arterial pressure (i.e., ~100 mmHg). Note the wall thinning and marked enlargement of both the left and right ventricular chambers induced by chronic volume overload secondary to an infrarenal AV fistula.

MMP activity. As can be seen in Fig. 5, LV MMP activity is significantly increased in rats with symptomatic CHF (27% greater than control, P < 0.01) compared with normal levels in rats from the compensated group.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Response of LV matrix metalloproteinase activity secondary to chronic AV fistula. To combine results from different gels, a single extract from the same control heart was used as a standard on all gels. The activity of the lytic bands in the other lanes of a gel were expressed as a percentage of this standard's activity. Once normalized in this fashion, the percent activities from hearts belonging to the same group (i.e., control, open bar; compensated, hatched bar; and failure, solid bar) were averaged. *P < 0.01 vs. CHF group. Values are means ± SD.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The cardiovascular remodeling secondary to chronic volume overload is characterized by progressive ventricular dilatation, hypertrophy, and ultimately failure. Although ventricular function in the clinically manifested stage of CHF has been well studied in both human subjects and experimental animal models (20, 31, 35, 37), very little is known about ventricular remodeling and function during the compensated period of chronic volume overload preceding the development of clinical heart failure. Such information would provide valuable insight into the pathogenesis of CHF. We previously established that the ventricular remodeling during the first 8 wk in a rat model of AV fistula consisted of progressive ventricular hypertrophy, dilatation, and increased compliance (5). Although there was some evidence of decreased contractility and developing CHF at 8 wk postfistula, overall, LV systolic function in the fistula groups was comparable to control. The findings from that study suggested that a ventricle subjected to a sustained volume overload never reaches steady-state compensation. Therefore, we performed the studies reported herein to complete the characterization of the functional effects of chronic ventricular remodeling and to determine whether these changes would eventually result in overt CHF. Our hypothesis was that ventricular dilatation without appropriate hypertrophy induces myocardial dysfunction and, therefore, is responsible for the transition to overt heart failure.

This is the first comprehensive study to document the temporal changes in ventricular mass, size, and function; myocardial MMP activity; and the interstitial collagen content associated with the compensated and decompensated stages of chronic volume overload induced by AV fistula. Furthermore, no other studies have previously documented the consistent development of overt CHF in the rat AV fistula model and characterized the temporal morbidity and mortality associated with CHF in this model. Accordingly, these results together with our previous article (5) provide guidance for investigators to design studies to examine specific molecular, biochemical, or cellular events relevant to different stages in the progression from compensated to decompensated heart failure. These findings also represent experimental confirmation of the hypothesis that the transition to heart failure is triggered by marked ventricular dilatation first put forward by Linzbach (21).

Hypertrophic response. Together with our earlier findings (5), this study establishes that myocardial hypertrophy induced by an AV fistula is progressive during the compensatory phase of chronic volume overload but appears to reach an upper limit before the development of CHF. This hypothesis is supported by no difference in the LV weights of both the compensated and failure groups. However, there does not appear to be a straightforward correlation between the degree of LV hypertrophy and development of overt CHF or death, because several rats that developed overt failure had relatively small LV weights (as little as 1,425 mg compared with a high value of 2,009 mg). A proportionally greater hypertrophic response may partially explain why particular hearts did not fail earlier, as suggested by the observation that the compensated phase of ventricular remodeling tends to have a longer duration in hearts with larger LV weights.

LV diastolic properties. Ventricular dilatation is the initial compensatory response to maintain an appropriate stroke volume in the volume-overloaded heart (16, 24). This time-dependent process of LV dilatation and associated changes in LV shape has been termed "remodeling" (31). We previously demonstrated that ventricular dilatation during the first 8 wk of postfistula is progressive. In addition to progressive dilatation, a significant increase in ventricular compliance developed between the third and fifth week of volume overload. However, as can be seen in Fig. 1, there was essentially no difference in ventricular size or compliance in the AV fistula groups beyond 8 wk when grouped on the basis of time postfistula. However, when these same animals were divided into groups on the basis of whether they were compensated or in heart failure, as depicted in Fig. 3, there was an appreciable increase in LV dilatation and compliance in the failing hearts with no additional increase in LV weight. These findings suggest that progressive ventricular hypertrophy associated with a compensated ventricle is limited, and additional size increases are detrimental. Linzbach (21) asserted that as the ventricle enlarges, a substantially greater force is necessary to eject an equivalent volume of blood. This adverse effect of dilatation on ventricular function is compounded by the fact that the increased ventricular size produces LV wall thinning, effectively requiring the amount of work per unit of contracting muscle to be still greater. Linzbach's hypothesis is supported by the finding by Gulch and Jacob (13) that stroke volume increases with growing anatomic heart size but ultimately reaches a maximum beyond which it rapidly declines. Thus this marked ventricular dilatation is potentially the primary impetus for the development of overt CHF.

Histological evaluation. With the exception of Z band misalignment, the histological appearance of the myocardium has been reported to be relatively normal in the AV fistula model (14, 30, 33). A report by Hatt et al. (14) found that degenerative changes identified at earlier stages of remodeling had largely resolved after 6 mo of chronic volume overload. Thus our findings are overall in agreement with the existing literature, even as to the normal appearance of the myocardium at the light microscopic level in rats with overt CHF.

MMP activity. The rats with symptomatic CHF have a significant increase in myocardial MMP activity, which has not previously been observed in hearts with marked remodeling that have not yet decompensated (26). This in itself is not a dramatic finding, given the several studies that have previously demonstrated that experimentally induced myocardial collagen degradation leads to ventricular enlargement and an increase in ventricular compliance (7, 19, 28). This increase in MMP activity seen in the failing hearts corresponds to a period of time when marked changes in LV dilatation and compliance were seen. Although this does not definitively prove that collagen degradation is responsible for the increases in chamber volume and compliance, it does support that hypothesis. However, these results also offer the first clear experimental evidence that significant ventricular dilatation can occur even when collagen synthesis exceeds degradation. This observation lends credence to the hypothesis that factors other than total collagen concentration, such as decreased collagen interactions with integrin receptors, altered collagen cross-linking, or internal architectural changes in the cardiomyocyte, are responsible for changes in ventricular compliance. Given the increases in myocardial MMP activity, it is likely that changes in the composition and organization of the collagen matrix occur even if the total percentage of collagen is maintained or increased (i.e., decreases in collagen cross-linking and/or alterations in the ratio of collagen type I to type III). New collagen fiber deposition in the heart tends to consist predominately of the more compliant type III collagen. The greater compliance of this new collagen, together with less extensive collagen cross-linking, could account for much of the progressive increase in ventricular compliance over the course of chronic volume overload in this model. Similarly, a decrease in integrin-mediated cardiomyocyte adhesion to the extracellular matrix could also account for increased ventricular compliance that would not be dependent on changes in interstitial collagen in the myocardium. Finally, the fact that marked LV dilatation occurred without a concomitant increase in LV weight suggests other mechanisms, such as alterations in the arrangement of muscle layers in the ventricular wall or intracellular remodeling in cardiomyocytes, contribute to the pathogenesis of heart failure. This might consist of the sliding displacement (i.e., slippage) of heart muscle cells leading to a decrease in the number of muscle layers in the ventricular wall proposed by Linzbach (22). However, even more likely is the contribution of myocyte lengthening to chamber dilatation. Intracellular reorganization consisting of in series sarcomeric addition has been shown to be a major component of ventricular remodeling secondary to chronic volume overload (23), although in these compensated hearts, Liu et al. found that increased cross-sectional area accounted for more of the increase in myocyte volume (i.e., hypertrophy) than cell lengthening. From our observations, however, we hypothesize that in the decompensated heart there is a progressive reduction in myocyte diameter (i.e., loss of myofilaments) together with further cellular elongation, which could account for the marked ventricular dilatation and wall thinning without a concurrent increase in ventricular weight. This hypothesis is consistent with the histological changes reported herein for the failing hearts. In addition, this intracellular remodeling is also likely to be a major factor contributing to myocardial dysfunction and decreased contractility associated with the development of CHF.

LV systolic function. The slope of the linear P_max-V relationship in the isolated canine heart has been shown to be sensitive to pharmacological changes in contractility, independent of preload and afterload, and therefore is considered to be an index of contractility (36). That is, positive and negative inotropic agents result in larger and smaller slope values, respectively. The results herein indicate the slope of the P_max-V was decreased for all of the AV fistula groups, suggesting that the intrinsic myocardial contractility was significantly depressed relative to control. The relatively greater decrease observed in the P_max-V relationship of the compensated 15-wk fistula group seemingly indicates that these rats were closer to developing CHF than the 8- and 21-wk compensated groups, which had P_max-V values that were not significantly different from the control group. This interpretation is also supported by the normal body weight of the 21-wk compensated group, indicating an absence of ascites. Accordingly, the 8- and 21-wk compensated rats had not progressed to the point of developing CHF and more than likely could have maintained compensated ventricular function beyond the study end point. The marked decrease in contractility observed in the fistula failure group was significantly greater (P < 0.01) than that seen in the compensated group; however, despite decreases in contractility, systolic function (i.e., peak isovolumetric pressure) was maintained, albeit at higher EDVs.

Morbidity and mortality. To date we have accumulated data on 116 rats postfistula, which illustrates the consistent development of overt CHF after an extended period of compensated volume overload (between 4 and 21 wk postfistula) in this model of chronic volume overload. The cumulative CHF-related morbidity and mortality in untreated fistulas at 21 wk postfistula is 80%. Furthermore, the morbidity and mortality study demonstrates that the development of overt CHF is not limited to a specific time interval following the imposition of a sustained volume overload. Although the extent of hypertrophy was highly variable in hearts that failed, the extent of LV dilatation was remarkably similar. Taken as a whole, the current findings suggest that the development of overt heart failure is triggered when the compensatory responses (i.e., hypertrophy, contractility, and Frank-Starling) are exhausted and marked ventricular dilatation is induced.

Possible mechanisms. Chronic volume overload consistently produces marked hypertrophy and ventricular dilatation regardless of the animal model; however, the specific mechanisms responsible for progressive ventricular dilatation, wall thinning, and sphericalization are still unknown. The traditional hypothesis is that ventricular systolic dysfunction leads to chamber dilatation; however, a more contemporary hypothesis is that development of overt systolic dysfunction is preceded by an adverse increase in chamber volume (9). Several recent investigations have emphasized that cardiac dilatation is a precursor of LV dysfunction and clinical heart failure (11, 32, 39). Our results are in agreement with these studies, clearly demonstrating that extensive myocardial remodeling producing marked ventricular dilatation and increased compliance precedes the development of overt CHF. However, stroke volume is generally increased by ventricular dilatation even if contractility is decreased (13). Therefore, cardiac output would be maintained until compensatory hypertrophic responses and contractility reserves were exhausted, leading to significant increases in EDP and additional ventricular dilatation. A possible contributing factor to the development of CHF could be compromised mitral valve integrity secondary to marked ventricular dilatation, with the resulting regurgitant flow further impairing cardiac output and increasing the degree of volume overload, inducing sudden decompensation and overt failure.