INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

GLYCOSAMINOGLYCANS (GAGs), and in particular their major component hyaluronan (2), are important for interstitial fluid volume (V_i) control. Hyaluronan contributes to the gel-like structure of the interstitium, defined as the space located between the capillary walls and the cells, and variation in the content and concentration of hyaluronan may affect interstitial fluid characteristics like hydrostatic pressure and volume. In addition, changes in GAG content may also be of importance in disease conditions affecting the connective tissue.

Because of the importance of GAGs in the regulation of tissue hydration, it is of interest to study the influence of changes in interstitial composition on V_i and some of its determinants in a hypothyroid state where such changes are likely to occur. In a recent study we addressed the transcapillary fluid balance in hypothyroid rats (23). We found that lack of thyroid hormones resulted in increased interstitial fluid pressure (P_i) in the skin and muscle and a reduced protein concentration in interstitial fluid. Hyaluronan was increased in the muscle and heart, but not in the skin and interstitial fluid from the skin and muscle. The increase in P_i was found despite a normal or even reduced V_i and suggests that the relationship between V_i and P_i, the interstitial compliance (C_i), is changed in the hypothyroid state. As stated by Aukland and Nicolaysen (1), C_i plays a central role in regulation of V_i because it determines the change in P_i resulting from a given change in V_i. The importance of compliance in volume regulation has recently been verified in mathematical modeling studies, showing that C_i is the single most important parameter in volume regulation (6, 24). Because our previous study suggested an altered C_i in the hypothyroid state, we decided to measure this parameter in the skin and muscle of thyroidectomized rats.

We found that there was a steeper volume pressure relationship and thus a reduced compliance during dehydration in the back skin and hindlimb muscle and that the increase in P_i that could be generated in overhydration, i.e., the hydrostatic counterpressure that can counteract filtration and thereby edema, was less in hypothyroid compared with euthyroid rats. These findings can partly be explained by an altered composition of the interstitium in the hypothyroid rat.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The experiments were performed in anesthetized female Wistar-Møller rats (207-279 g) fed a standard laboratory diet. While the rats were anesthetized, body temperature was maintained at 36.5-37.5°C with a heat lamp. The rats were not fasted before or during the experiments. All experiments were performed in accordance with recommendations given by the Norwegian State Commission for Laboratory Animals and were approved by the local ethical committee.

Surgical Procedures

To verify a hypothyroid state, thyroxin (T₄) was measured in plasma sampled at the end of the experiment in a routine assay at a local hospital. T₄ in the hypothyroid rats averaged 17.8 ± 3.5 nmol/l (SD, n = 17), whereas the corresponding value in euthyroid rats was 51.4 ± 14.5 nmol/l (SD, n = 19) (P < 0.001). Plasma Ca²⁺ averaged 2.72 ± 0.13 (SD, n = 7) and 2.86 ± 0.20 mmol/l (SD, n = 7) in hypothyroid and euthyroid rats, respectively (P > 0.05).

At the day of the experiment, the rat was anesthetized with pentobarbital (50 mg/kg ip). After tracheotomy, polyethylene-50 catheters were inserted in the left carotid artery for measurement of blood pressure and blood sampling and in the left jugular vein for injections.

Measurement of Interstitial Fluid Pressure

The rat was placed in a supine position, and micropipettes were inserted on the medial aspect of the thigh through intact skin, or in the case of muscle, in the fiber direction through the fascia exposed by a 2- to 3-mm long skin incision. P_i was also measured caudally in the back skin after the rat was placed in a prone position. The P_i measurements were accepted when: 1) pressure remained unaltered when feedback gain was changed, 2) communication between pipette and interstitial fluid could be demonstrated, and 3) zero pressure did not change during one measurement. P_i was taken as the mean of two to four pressure measurements in each tissue. Pressure measurements were performed during the first 2 h of isotope equilibration and after an equilibration period following induced changes in tissue hydration (see below).

Measurement of Distribution Volumes

After an equilibration period following the induced change in hydration (see Experimental Protocol), 3-4 µCi of ¹³¹I-HSA was injected and allowed 5 min for circulation. A final blood sample of 0.5-0.7 ml was obtained from the arterial catheter, and the rat was killed with an intravenous overdose of anesthetic.

Hair was carefully removed from the hind legs and lower back with fine clippers, and 0.3- to 0.5-g samples of skin and muscle corresponding to those taken in euvolemia were taken from the hindlimb not used for euvolemic control samples and of the back skin avoiding the area used for the euvolemia sample by minimum 2 cm. Tissue samples were placed in tared, covered vials and weighed. After counting was completed, tissue samples were dried at 60°C to constant weight (change < 1 mg in 24 h). Tissue water content (V_w) was the difference between wet and dry weights.

Samples were counted in a LKB gamma counter (model 1282 Compugamma) using window settings of 15-75 keV for ¹²⁵I, 290-350 keV for ⁵¹Cr, and 350-470 keV for ¹³¹I. Standards were counted in every experiment to obtain spillover corrections, and counts were corrected for background and spillover and for ¹³¹I radioactivity decay during the period of measurement.

Experimental Protocol

Dehydration. Dehydration was induced by peritoneal dialysis using a hypertonic glucose solution. With P_i measured and tissue sampled in euvolemia, 10-15 ml of 20% glucose in Ringer solution were instilled via a catheter introduced into the peritoneal cavity from the lumbar region. The osmolarity of the dialyzing fluid was about 1,400 mosml/l as determined by a vapor pressure osmometer (model 5500, Wescor). After an equilibration period of 45 min, the dialyzing fluid was withdrawn and the dialysis repeated. By this procedure a net volume of 18-29 ml was removed from each rat. During dialysis, repeated intravenous infusions of 2 ml 10% human serum albumin in Ringer solution (totaling 4-6 ml) were given in an attempt to prevent hypotension. Blood glucose measured in two hypothyroid and two euthyroid rats with a Heamo Glucometer rose from 4-5 mmol/l in euvolemic control situation to 30-40 mmo/l after the first dialysis and to 80-100 mmol/l after the second dialysis. The control level of glucose and level after dialysis did not differ between hypothyroid and euthyroid rats.

Overhydration. Previous studies of the volume-pressure relationship in rats suggest that there initially is a linear relationship between V_i and P_i at low degrees of overhydration followed by no increase in P_i as the overhydration increases. A gradual overhydration was therefore induced by volume loading. After measurements of P_i and sampling of tissue in euvolemia, acetated Ringer was infused corresponding to 5, 10, and 20% of the body weight. The infusion lasted 15, 30, and 60 min, respectively. Measurement of P_i started 90 min after ended infusion and was completed within 60 min. With the pressure measurements being finished, ¹³¹I-HSA was injected for plasma volume determination, and plasma and tissue were sampled for volume measurements as described above.

Time control. Because the same dose of ⁵¹Cr-EDTA was used for measurement of extracellular volume in the euvolemia and experimental situation, an additional group of four hypothyroid and four euthyroid rats was included to see whether the distribution volume increased with the duration of the experiment. In this group, ⁵¹Cr-EDTA was injected after kidney ligation, and blood and tissue samples were obtained after injection of ¹²⁵I-HSA for plasma volume determination corresponding to the euvolemia equilibration period of 120 min. After a period of 3.5 h where no fluid was given, ¹³¹I-HSA was then injected and allowed 5 min for circulation before the rat was killed and blood and tissue were sampled for volume determination as described above.

Calculations

(1)

(2)

(3)

C_i was calculated from the relationship between V_i and P_i in two different ways. First, this compliance was calculated by conventional linear regression analysis on euvolemia and dehydration data (least squares method). Data from this analysis were used in the analyses of covariance (see Statistics). In regular regression analysis it is assumed that there are errors only in the independent variable, and therefore the coefficient of regression will be depending on which parameter is chosen as the independent variable. The regression coefficients are therefore given as the geometric mean obtained when using V_i as the independent as well the dependent parameter in the regression analysis (4).

Statistics

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

At the time of the thyroidectomy, experimental and euthyroid rats weighed 237 ± 14 (n = 17) and 235 ± 10 g (n = 19), respectively (P > 0.05). Lack of thyroid hormone affected body weight. Rats used for experiments 8 wk after thyroid extirpation weighed 234 ± 10 g compared with 270 ± 21 g in euthyroid rats (P < 0.001).

At the day of the experiment rectal temperature measured as soon as possible after induction of anesthesia averaged 36.8 ± 0.6 and 38.1 ± 0.8°C (P < 0.01) in hypothyroid and euthyroid rats, respectively, with corresponding mean arterial blood pressures of 69 ± 10 and 105 ± 17 mmHg (P < 0.001), respectively. This difference in arterial pressure may affect fluid distribution in the tissues but will not influence the compliance, which is calculated from directly measured V_i and P_i.

Interstitial Fluid Volume and Pressure in Euvolemia

Whereas P_i was consistently more positive in hypothyroid than in euthyroid rats, less consistent differences were found for fluid volumes in these two conditions. In the hindlimb skin, V_i and V_w did not differ significantly (data not shown), whereas these volumes were significantly lower in the back skin of hypothyroid rats compared with euthyroid rats, V_i averaging 0.409 ± 0.059 and 0.437 ± 0.051 ml/g wet wt and V_w 0.638 ± 0.031 and 0.684 ± 0.013 ml/g wet wt for hypothyroid and euthyroid rats, respectively (P < 0.01 for both comparisons).

In contrast to the back skin, volumes in the hindlimb skeletal muscle were higher in hypothyroid than in euthyroid rats. Thus V_i averaged 0.110 ± 0.013 and 0.099 ± 0.009 ml/g wet wt and V_w 0.814 ± 0.009 and 0.793 ± 0.011 ml/g wet wt, respectively (P < 0.01 for both comparisons).

Interstitial Fluid Volume and Pressure During Altered Hydration

As might be expected, there was a gradual increase in V_i and V_w in all the tissues studied. Thus, after infusion of Ringer solution, 20% of body weight, average V_i in hypothyroid rats was 58, 51, and 120% of the volume in euvolemia in hindlimb skin, back skin, and skeletal muscle, respectively, with corresponding numbers in euthyroid rats of 72, 52, and 127%.

Dehydration led to mean reductions in P_i of 2-3.5 mmHg in the tissues studied. None of the mean pressures obtained in hypothyroid rats differed significantly from those in corresponding tissues in euthyroid rats.

Peritoneal dialysis led to reduced V_i and V_w in all tissues and both categories of rats. Of the V_i values, only the V_i of 0.072 ± 0.008 ml/g wet wt in skeletal muscle of hypothyroid rats differed significantly from the corresponding value of 0.065 ± 0.014 ml/g wet wt (P < 0.01) in euthyroid rats.

Interstitial Fluid Volume-Pressure Relationship

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Interstitial fluid pressure (Pi) and volume (Vi) in hindlimb skin. A: corresponding absolute values of Vi and Pi in euthyroid (circles) and hypothyroid rats (triangles). Filled symbols, euvolemia; open symbols, situation after saline expansion or dehydration induced by peritoneal dialysis. B: changes in Pi and Vi from euvolemia to experimental situation in hypothyroid (open circles) and euthyroid rats (closed circles) calculated from data shown in A. +Position of euvolemia Pi and Vi. Solid (euthyroid) and dashed (hypothyroid) lines in dehydration calculated as the geometric mean of regression lines (see METHODS) and drawn to best visual fit in overhydration.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Pi and Vi in back skin. A: corresponding absolute values of Vi and Pi in euthyroid (circles) and hypothyroid rats (triangles). Filled symbols, euvolemia; open symbols, situation after saline expansion or dehydration induced by peritoneal dialysis. B: changes in Pi and Vi from euvolemia to experimental situation in hypothyroid (open circles) and euthyroid rats (closed circles) calculated from data shown in A. +Position of euvolemia Pi and Vi. Solid (euthyroid) and dashed (hypothyroid) lines in dehydration calculated as the geometric mean of regression lines (see METHODS) and drawn to best visual fit in overhydration.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Pi and Vi in hindlimb skeletal muscle. A: corresponding absolute values of Vi and Pi in euthyroid (circles) and hypothyroid rats (triangles). Filled symbols, euvolemia; open symbols, situation after saline expansion or dehydration induced by peritoneal dialysis. B: changes in Pi and Vi from euvolemia to experimental situation in hypothyroid (open circles) and euthyroid rats (closed circles) calculated from data shown in A. +Position of euvolemia Pi and Vi. Solid (euthyroid) and dashed (hypothyroid) lines in dehydration calculated as the geometric mean of regression lines (see METHODS) and drawn to best visual fit in overhydration.

Hindlimb skin. Data for hindlimb skin are shown in Fig. 1. As stated above, the lowest degree of overhydration resulted in a marked rise in P_i, whereas a further volume rise did not lead to further rise in P_i. The average increase in P_i in hypothyroid rats of 0.4 ± 0.5 mmHg was significantly lower than the corresponding value in euthyroid rats of 1.1 ± 0.6 mmHg (P < 0.05) (n = 9 for both types of rats).

1

1

Back skin. Figure 2 shows the volume-pressure data for back skin. As for hindlimb skin, there was a marked rise in P_i at a modest rise in V_i, and a further rise in V_i did not lead to an increase in P_i. When the above data were grouped, we found that the average increase in P_i in overhydration was 0.6 ± 0.2 mmHg in hypothyroid and 1.2 ± 0.3 mmHg in euthyroid rats (P < 0.01).

1

1

1

1

Hindlimb muscle. Figure 3 shows the volume-pressure data for hindlimb muscle. In the muscle there was a less abrupt rise in P_i in the initial phase of overhydration than observed in the hindlimb and back skin. As for the other tissues, a further increase in volume did not lead to an increase in P_i. In overhydration P_i averaged 0.4 ± 0.5 and 1.1 ± 0.6 mmHg in hypothyroid and euthyroid rats, respectively (P < 0.05).

1

1

1

1

Time control. The respective distribution volume of ⁵¹Cr-EDTA and the measured V_i did not differ significantly after 2 and 5.5 h of circulation time for any of the tissues studied in hypothyroid as well as euthyroid rats.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hypothyrosis is an important clinical condition, and we have shown for the first time that the structural changes associated with this condition lead to changes in the V_i-P_i relationship. We verified that the P_i in euvolemia (control situation) is increased in hypothyrosis (23). In addition, we found that the rise in P_i and thereby the hydrostatic counterpressure generated in overhydration is less in hypothyroid than in euthyroid rats. Furthermore, hypothyroid skin and muscle were found to have a lower C_i in dehydration that may be a result of an altered composition of the interstitium and may have consequences for body fluid balance. We will first discuss some methodological aspects and then compare our results with previous data and finally discuss the physiological implications of the present data.

Methodological Aspects

To get euvolemic and experimental values in the same animal, we used hypertonic glucose to produce acute dehydration, a procedure that caused massive hyperglycemia and a hypertonic dehydration. In addition fluid mobilization from the tissues was augmented by infusion of hyperoncotic human serum albumin. The method of dehydration may influence the calculated compliance, but previous experiments have shown that compliance in the skin and muscle is the same whether dehydration is produced isotonically or hypertonically (21, 26).

Comparison With Previous Studies

To our knowledge there are no previous studies on compliance in hypothyrosis. Whereas there were important quantitative differences between the curves, which will be addressed below, the overall shape of the volume-pressure curve was similar in hypothyroid and euthyroid rats, i.e., a linear relationship between V_i and P_i in dehydration and in the initial phase of overhydration, whereas an increase in V_i 20% above control did not lead to a further rise in P_i. This curve outline also agrees well with that obtained in pioneer studies by Guyton (9) and in later studies in our and other laboratories (3, 25, 26) (for older references see Ref. 2).

Shape of Volume-Pressure Curves

Of interest in this connection is also the significant rise in P_i at the smallest volume expansion of 5% of body weight with a tendency to falling pressures at a larger expansion creating a "bump" on the volume-pressure curve. Although our number of experiments in overhydration is low, these data correspond well to observations in the lung where induction of a small rise in wet-to-dry ratio resulted in an abrupt rise in P_i from 10 to +3 cmH₂O, whereas with a further expansion the pressure leveled off around zero with no tendency to rise at almost a doubling of the wet-to-dry ratio (14), i.e., a volume-pressure relationship in overhydration very similar to ours. The authors suggested that the reduction in P_i in progressing overhydration resulted from a degradation of proteoglycans and weakening of the bonds between the structural elements of the extracellular matrix. The similarity of the volume-pressure curves suggest that this may also have occurred in our study, but we have no data to support that assumption.

Compliance in dehydration differed depending on the organs studied, and these will be considered separately. In the muscle, C_i in hypothyrosis was only 50% of that in euthyroid rats, and V_i and total tissue water were significantly increased, suggesting a substantial change in interstitial fluid dynamics. The explanation for this discrepancy in C_i may be found considering the structure of the interstitium in hypothyrosis. In a previous study we found that hyaluronan was 56% above control in hypothyroid muscle tissue. The hyaluronan molecule is long chained and entangled, strongly negatively charged at physiological pH, resulting in mutual repulsion forces within the molecule (7). It is therefore not unlikely that the increased hyaluronan content in hypothyroid muscle will lead to an increased gel osmotic pressure and thereby a more negative P_i for the same change in V_i in hypothyroid than in euthyroid rats, i.e., a reduced compliance in the former as observed.

Whereas an increased hyaluronan content may be the explanation for the reduced compliance in hypothyroid muscle, this substance cannot be the reason for the corresponding C_i reduction in back skin, because earlier studies have shown a similar content in this tissue in hypothyroid and euthyroid rats (23). Furthermore, analyses showed that the content of uronic acid, an indicator of total glycosaminoglycans, was similar in the two situations as well. It is possible that the hydration at the outset of the dehydration, i.e., in the euvolemic control situation will influence compliance. As observed previously (23), V_i in back skin was about 10% lower in hypothyroid compared with that in euthyroid rats. The exact reason for this reduced volume is not obvious, but the influence of reduced thyroid hormones on cell volume regulation may be of importance (8). A change in the structural elements may also have contributed to this observation. Lack of thyroid hormones influences collagen metabolism (16), and a reduced collagen content was recently found in the back skin of hypothyroid rats (23). Furthermore, in the myocardium of hypothyroid rats, Klein and co-workers (11) found significant remodeling of collagen shown by an increased thickness of individual type I collagen fibers, which may influence the organization of the collagen matrix of the interstitium and thereby influence compliance.

In the hindlimb skin, compliance was similar in euthyroid and hypothyroid rats. Furthermore, fluid volumes did not differ in the two types of animals, as observed for this tissue in a previous study (23). It is surprising to find differences from control in compliance in some tissues and not in others considering that we are looking at a hormonal effect, which may be expected to affect all tissues in the same direction. To give a definite answer we would have to know in more detail what determines compliance. As shown by the data from the back skin, hyaluronan is not the only determinant of C_i, but the affected cell volume regulation (8) and the structural changes in collagen mentioned above (11) may also be of importance resulting in an effect of thyroid hormones on C_i differing between tissues.

Physiological Implications of a Changed Compliance

These data permit some potential estimates of available fluid (i.e., "reserve volume") for skin and muscle if P_i is lowered, i.e., in dehydration. If capillary pressure is reduced by 1 mmHg and all other Starling forces remain constant, the P_i would reach a new level after removal of 0.65 and 0.94 ml/100 g rat from skin of hypothyroid and euthyroid rats, respectively. The corresponding numbers for skeletal muscle are 0.59 and 0.91 ml/100 g rat. Although recent data suggest less influence of interstitial fluid proteins in fluid transport than previously anticipated (10, 12), it is likely that this fluid removal will be opposed slightly by an increased concentration of interstitial proteins and that P_i has to be reduced more to mobilize the actual volumes. Still, these calculations suggest that for skin and muscle close to 50% more fluid may be mobilized to the circulation in euthyroid than in hypothyroid rats when P_i is reduced by 1 mmHg.

In summary, we have shown that induction of hypothyrosis lead to changes in the relationship between V_i and P_i in back skin and skeletal muscle during perturbations in V_i. P_i in euvolemia is higher and the increase that can be induced by overhydration lower in hypothyroid than in euthyroid rats, suggesting that a rise in P_i is less important as an edema preventive mechanism in hypothyrosis. C_i in dehydration is reduced in the back skin and muscle in hypothyroid compared with euthyroid animals, which may result in a reduced ability to mobilize fluid from the interstitium in hypothyrosis and thereby a lower reserve volume in dehydration.