Skip Navigation

Cardiovascular Research 1997 33(1):230-240; doi:10.1016/S0008-6363(96)00198-8
© 1997 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Heron, M. I.
Right arrow Articles by Rakusan, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heron, M. I.
Right arrow Articles by Rakusan, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1997, European Society of Cardiology

Early and late effect of neonatal hypo- and hyperthyroidism on coronary capillary geometry and long-term heart function in rat

Marcia I. Herona, Frantisek Kolarb, Frantisek Papousekb and Karel Rakusana,*

aDepartment of Physiology, Faculty of Medicine, University of Ottawa, 451 Smyth Road, Ottawa, Ont., Canada K1H 8M5
bInstitute of Physiology, Academy of Sciences of the Czech Republic, CZ-142 20 Prague, Czech Republic

Received 26 March 1996; accepted 10 August 1996


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objective: The aim of the present study was two-fold: (1) to examine the effect of hyper- and hypothyroidism on the developing coronary capillary network in neonatal rats, and (2) to determine in adult rats that had re-established euthyroid status whether long-term changes in capillary geometry or cardiac function had been induced by either neonatal thyroid condition. Method: Two-day-old rats were treated every other day for 12 or 28 days with either 3,3',5-triiodo-l-thyronine or 0.05% 6-n-propylthiouracil. After this time, treatment was stopped and in two-thirds of the rats morphometric examination of capillary geometry and immunohistochemical detection of proliferating cell nuclear antigen (PCNA) expression in endothelial cell nuclei were conducted. Remaining rats were weaned and grew to 80 days of age, at which time persistent changes in capillary geometry, PCNA expression, and cardiac function were assessed. Results: Neonatal hyperthyroidism induced cardiomegaly (P < 0.01), whereas neonatal hypothyroidism attenuated cardiac growth (P < 0.01). Capillary numerical density, capillary segment lengths, and PCNA-labelling analysis indicated marked capillary growth in hyperthyroid rats (P < 0.05), but attenuated capillary growth in hypothyroid rats. The elicited capillary growth response appeared to be more dependent on altered tissue maturation than on cardiac growth rate. After discontinuing treatment both neonatal thyroid conditions induced a deficit in left ventricular growth (P < 0.01). Furthermore, neonatal hyperthyroidism appeared to inhibit subsequent capillary growth in distal regions of the capillary bed in addition to inducing lasting positive chronotropic and inotropic effects on cardiac function (P < 0.05). Neonatal hypothyroidism did not produce any lasting changes in capillarization or in cardiac function. Conclusions: Results suggest that neonatal thyroid status influences early growth and development of the coronary capillary network, possibly by regulating tissue maturation, as well as inducing lasting effects on subsequent cardiac and capillary growth and heart function.

KEYWORDS Hypothyroidism; Hyperthyroidism; Coronary vasculature; Myocardial function; Capillary; growth; Angiogenesis; Development; Hypertrophy; Lectin; Morphometry; PCNA


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
A close relationship between thyroid status, cardiac growth, and coronary capillary growth has been well-established in adult mammals. Adult-onset hyperthyroidism induces a sizeable increase in cardiac mass with proportional [1, 2] or excess [3] coronary capillary growth. In comparison, adult-onset hypothyroidism is associated with cardiac atrophy and some degree of coronary capillary growth [2, 4]. Based on these findings, one may expect the effect of increased or decreased thyroid hormone levels on the coronary capillary network would be even more pronounced during the neonatal stage of cardiac development, when the microvascular bed is undergoing considerable growth and expansion [5, 6]. In fact, studies comparing the effect of pressure overload on coronary adaptations in neonatal and adult mammals have reported that the cardiac response elicited in neonates is comparably different from the response in adults [7, 8]. Few studies exist, however, which have examined the response of the developing coronary capillary network to altered neonatal thyroid status.

Thyroid hormones are known to play an important role in normal growth during the neonatal period of development [9]. As in humans, neonatal hypothyroidism in the rat leads to delayed maturation [10], whereas the opposite has been proposed for hyperthyroidism in the rat [11]. Studies on cardiac growth and muscle cell proliferation in rats have shown that treatment with thyroid hormone during the neonatal period of development not only leads to accelerated growth of the heart, but also influences cardiac muscle cell proliferation and consequently cardiac muscle cell numbers [12–14]. The neonatal period in the rat is marked by substantial proliferation and expansion of the coronary capillary network. In fact, close to one half of all the capillaries present in the adult rat heart are formed during the first 3–4 postnatal weeks [15]. After this time, only a few new capillaries are formed as the rat grows into adulthood. Thus, as in the case of cardiomyocytes, alterations in plasma thyroid hormone levels during the neonatal period may also affect the normal formation and maturation of the coronary capillary network. Since the geometrical arrangement of the capillary network is an important determinant of the diffusion conditions for oxygen transfer to myocardial tissue, irregular formation of the microvascular bed may result in inadequate myocardial oxygenation, possibly leading to compromised cardiac development. In turn, abnormal growth of the heart may have a long-term positive or negative impact on cardiac function.

The aim of the present study was, first, to examine morphometrically the effect of early neonatal-onset hypo- and hyperthyroidism on the growth and development of coronary capillaries, and, second, to determine in adult rats whether long-term changes in capillary geometry or cardiac function had been induced by either neonatal thyroid condition. Furthermore, because altered neonatal thyroid status influences the rate of maturation, this model and results of this study will be useful in determining if the early, rapid formation of new capillaries and subsequent decline is dependent on the rate of cardiac growth or tissue maturation.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Experimental design
Newborn male and female Sprague-Dawley rats (Charles River) were used in this study. On the first postnatal day, all newborn rats were removed, mixed, and randomly redistributed to the dams in order to minimize any confounding effect of genetics or litter size on early postnatal development. Each litter, now consisting of approximately 10 rats per dam, was randomly assigned to one of 3 treatment groups: control (CON), hyperthyroid (HYPER), or hypothyroid (HYPO). Treatment began on the second postnatal day according to the following schedule: CON rats received subcutaneous (s.c.) injections of 0.9% saline (0.1 ml/100 g body mass) every second day; HYPER rats received s.c. injections of 3,3',5-triiodo-l-thyronine (T3; 20 µg/100 g body mass) every second day; and HYPO rats had 0.05% 6-n-propylthiouracil (PTU) added daily to the drinking water of the mothers; PTU is effectively transferred to nursing rats through the mothers' milk [13]. The preceding treatment protocol is similar to the one used by Kolar and colleagues [11].

On postnatal day 12, rats in one third of the litters from each of the 3 treatment groups were sacrificed (CON-12, n = 11; HYPER-12, n = 8; HYPO-12, n = 9). In remaining litters, treatment continued until postnatal day 28. At this time rats in half of the litters from each treatment group were sacrificed (CON-28, n = 7; HYPER-28, n = 10; HYPO-28, n = 9), while remaining rats were weaned and allowed to grow until approximately 80 days of age (CON-80, n = 8; HYPER-EU-80, n = 9; HYPO-EU-80, n = 9). Body mass was monitored in neonatal rats during the treatment period. Mothers and weaned rats were maintained on standard rat diet ad libitum. Though both male and female newborn rats were treated, only the hearts from male rats were used in subsequent morphometric analysis; female rats were used to obtain blood samples for serum T3 analysis. This investigation conforms with the Guide for Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1985).

Development of hypo- and hyperthyroidism was verified by radioimmunoassay analysis of total serum T3 levels (Diagnostic Veterinary Service, Ontario, Canada). Blood samples were collected from halothane-anaesthetized female rats, either by decapitation and bleeding (12-day-old rats) or by cardiac puncture (28- and 80-day-old rats). Samples were pooled within experimental groups to obtain sufficient serum for testing.

2.2. Functional measurements
Heart rates were determined in 12- and 28-day-old halothane-anaesthetized male rats from an ECG recording on a Grass 1P1F Polygraph. Blood pressure in the left (LV) and right (RV) ventricles of 80-day-old male rats, anaesthetized with pentobarbital sodium (52 mg/100 g body mass, i.p.), was measured using a Millar catheter-tip transducer connected to the Polygraph and a PC computer. The catheter was inserted into the left ventricular cavity via the right carotid artery under continuous pressure monitoring. Measurements were recorded after a stabilization period of 10–15 min; recordings were made every 5 min over a 15–20 min interval and the mean was calculated. Similarly, the catheter was introduced into the RV via the right jugular vein.

The analog pressure signal was digitized with a sampling frequency of 1 kHz and stored on computer for later processing. The following parameters were derived: systolic pressure (LVSP, RVSP), left ventricular end-diastolic pressure (LVEDP), left ventricular developed pressure (LVDevP), and the maximal rates of pressure development (+ dP/dt)max and fall (– dP/dt)max. In addition, the time constant of relaxation ({tau}) was calculated on the basis of an experimental model of isovolumic pressure decay, as the time required for the pressure at (– dP/dt)max to be reduced by 1/e [16]. Heart rate for 80-day-old rats was calculated from the LV pressure signal.

2.3. Histologic preparation
Under anaesthesia, hearts from male rats were arrested in diastole using a bolus injection of concentrated KCl. After being excised, surrounding fat and fibrous tissue were trimmed away and the atria were separated from the ventricles. The right ventricular free-wall was dissected free and weighed. The intact left ventricle and septum were also weighed and sectioned into 2 (12-day-old) or 3 (28- and 80-day-old) portions, perpendicular to the base-apex axis of the heart. Apical portions were used for percent tissue dry weight estimations, while remaining portions were snap-frozen in liquid nitrogen and stored at –80°C until further analysis. Only the mid-myocardium of the left ventricular free-wall was used for morphometric analysis.

Frozen tissue sections (3 sections per heart) were processed using 3 different histological methods: Bandeiraea simplicifolia I (BSI) lectin, alkaline phosphatase/dipeptidyl peptidase IV (AP/DPP IV), or proliferating cell nuclear antigen (PCNA) immunocytochemistry. The lectin method uses a specific lectin (BSI) to probe the terminal {alpha}-galactosyl saccharides associated with endothelial cell surfaces. In conjunction with a visible marker such as a peroxidase derivative, one can localize capillaries within tissue cross-sections [17]. The advantages of lectin over other capillary staining methods are: (1) that capillaries can be readily identified in tissue cross-sections irrespective of their perfusion state or degree of enzyme activity[18] and, more importantly for this study, (2) the sites for BSI lectin binding are present in the capillary bed of developing hearts and are stable during hypoxia [19], making this a reliable method for quantification of the capillary bed in immature hearts under different experimental conditions.

AP/DPP IV staining is based on the differential localization of alkaline phosphatase and dipeptidyl peptidase IV enzymes within the capillary walls. Alkaline phosphatase is located predominantly in proximal regions of the capillary network (i.e., in capillaries close to their feeding arteriole) whereas dipeptidyl peptidase IV is localized predominantly in distal regions of the capillary network (i.e., in capillaries close to their collecting venule). This method was used in longitudinal sections of mid-myocardium to distinguish between proximal and distal capillaries. However, since this method relies on functional enzymes, it is not applicable in the hearts from rats younger than 21 days of age [6], and therefore only 28 and 80-day-old hearts were subject to this process. Longitudinal analysis of the capillary network is important for a comprehensive picture of capillary geometry as changes along the longitudinal axis (e.g., capillary elongation) may not be detected by cross-sectional analysis alone. The AP/DPP IV method also has the advantage of enabling localization of changes in capillary geometry to proximal or distal portions of the capillary network.

Finally, immunocytochemical detection of PCNA protein in combination with haematoxylin and eosin staining[20] enables identification and quantitation of the different myocardial cell types that are within the synthesis phase of the cardiac cycle. PCNA data were used to corroborate morphometric data.

2.3.1. Bandeiraea simplicifolia I lectin staining
Tissue cross-sections (12 µm) from 12-, 28- and 80-day-old hearts were pre-incubated in phosphate-buffered saline (PBS; pH 7.4) (10 min, room temperature) before exposure to peroxidase-conjugated BSI lectin (25 µg/ml in PBS; Sigma) for 30 min at room temperature. Sections were then incubated with 3'-diaminobenzidine (DAB) solution (DAB Kit; Dimension Laboratories Inc.; 30 min), rinsed with PBS, and cover-slipped. Capillary profiles were identified in cross-sections by their brown color.

2.3.2. Alkaline phosphatase/dipeptidyl peptidase IV staining
For a detailed description of the AP/DPP IV histochemistry and protocol please, see Lojda [21] and Batra et al. [22]. Longitudinal sections (16 µm) from 28- and 80-day-old hearts were initially fixed in chloroform/acetone, and then incubated in a medium containing a substrate for the DPP IV enzyme (90 min, room temperature). Sections were rinsed with distilled water, and subsequently incubated in a medium containing a substrate for the AP enzyme (25 min, room temperature). After rinsing sections were cover-slipped. Proximal capillaries were identified in longitudinal sections by the resulting blue color, whereas distal capillaries were identified within the same sections by their red color.

2.3.3. PCNA/haematoxylin and eosin staining
Cross-sections (6 µm) from 12-, 28, and 80-day-old hearts were fixed in acetone and then incubated in 3% hydrogen peroxide/methanol. A primary mouse monoclonal anti-PCNA antibody (DAKO-PCNA, PC10; Dimension Laboratories Inc.) was then applied to sections. After rinsing in PBS, a secondary goat anti-mouse IgG2a (gamma-specific) peroxidase-conjugated antibody (Cedar Lane) was applied. Sections were incubated with a DAB solution (DAB Kit; Dimensions Laboratories Inc.), stained with haematoxylin and eosin and cover-slipped. PCNA-positive nuclei were identified in cross-sections by their brown color. For a more complete description the reader is referred to Heron and Rakusan [20].

2.4. Tissue dry weight determination
Apical portions were initially weighed (wet weight), dried in an oven (60°C) for more than 36 h, and re-weighed (dry weight). The percent of tissue dry weight is represented by the ratio of dry weight to wet weight.

2.5. Morphometric analysis
2.5.1. BSI-lectin-stained cross-sections
Five or 6 drawings demarcating the location of capillaries within a tissue cross-section were made for each heart, using an Olympus microscope and drawing arm attachment. As BSI lectin stains arterioles and venules as well as capillaries, tissue areas containing larger, easily identifiable arterioles and venules were not assessed. However, very small arterioles and venules that are similar in size to expanded capillaries were included. The average capillary numerical density (number of capillaries per unit area) was calculated for each heart.

2.5.2. AP/DPP-stained longitudinal sections
The length of capillary segments, defined as the distance of a capillary portion located between two sequential bifurcations [23], was measured and recorded as a function of capillary type (i.e., proximal or distal). A total of approximately 150 capillary segments were measured per heart using an Olympus microscope attached to an IBM-computer-based image analysis program (Bioquant, R and M Biometrics). Only segments which could be traced in their entirety from one branch point to another were measured. ‘H-type’ connecting segments were not included in the analysis.

2.5.3. PCNA/H&E-stained cross-sections
Approximately 600 nuclei were counted per heart (magnification 630 x) and categorized as endothelial (ENDO), myocyte (MYO), or ‘other’ nuclei (OTHER). The OTHER category included all nuclei that could not be identified as either ENDO or MYO. Nuclei were identified as ENDO if they protruded into the capillary lumen or were clearly part of the cells comprising the capillary. If these criteria were not met, the nuclei were categorized as OTHER. When the nucleus type was questionable, a higher magnification (1000 x) was used for identification. PCNA-immunoreactive nuclei were identified and categorized as PCNA-positive endothelial (ENDO +), PCNA-positive myocyte (MYO +), or PCNA-positive ‘other’ (OTHER +) nuclei. The number of PCNA-positive nuclei within each category was determined and expressed as a percentage of the total nuclei counted. The overall nuclei density (number of nuclei per unit area) and relative proportion of ENDO, MYO, and OTHER nuclei were estimated based on nuclei counts from 2–4 fields (23 250 µm2 per field).

2.6. Statistical analysis
Results are presented as mean ± s.e.m. Body and heart mass data, capillary numerical density and segment length data were analyzed using a two-way ANOVA with differences among group means determined using the Tukey HSD post hoc test. Segment length data demonstrated a logarithmic distribution and consequently were logarithmically-transformed prior to statistical analysis. The logarithmic transformation, which was determined to be the most appropriate (residual analysis), results in data that more closely follow a normal distribution and can thus be tested using an ANOVA.

Chi-square ({chi}2) analysis was used to determine overall significant differences in nuclei proportions among the groups. PCNA data were transformed using an arcsin transformation and then the transformed data were analyzed using a two-way ANOVA and Tukey HSD post hoc test. All results were considered statistically significant at P < 0.05.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
3.1. Serum T3 levels and heart rate measurements
Development of hyper- and hypothyroidism after 12 (CON-12, 0.7 nmol/1; HYPER-12, 1.3 nmol/l; HYPO-12, < 0.2 nmol/l) and 28 days (CON-28, 0.4 nmol/l; HYPER-28, 23.0 nmol/l; HYPO-12, < 0.2 nmol/l) of T3 treatment were confirmed by serum T3 analysis. Corresponding changes in heart rate in 12- and 28-day-old rats in the HYPER or HYPO groups agreed with the results of the serum T3 analyses (Fig. GR1). Rats in the HYPER groups had a higher heart rate, which only reached statistical significance after 28 days of treatment (P < 0.05), whereas HYPO rats had significantly lower heart rates at both 12 and 28 days (P < 0.01) compared to age-matched controls.


Figure 1
View larger version (42K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR1 Heart rates in 12- and 28-day-old control (CON), hyperthyroid (HYPER), and hypothyroid (HYPO) rats, and in rats 52 days after treatment was stopped (80 days old). Heart rate increased significantly between 12 and 28 days in CON and HYPER rats, and between 28-day-old HYPO and 80-day-old previously HYPO rats, at P < 0.01. Values are means, error bars = s.e.m. {delta} P < 0.01 vs. CON; * P < 0.05 vs. CON; {psi} P < 0.01 vs. HYPER.

 
At 80 days of age, 52 days after treatment had stopped, serum T3 levels were comparable among the groups (approximately 1.0 nmol/l), suggesting that euthyroidism had been re-established in neonatal HYPER and HYPO rats. Heart rates were similar between 80-day-old control and previously HYPO rats in accordance with T3 levels; however, the heart rate in previously HYPER rats remained higher than control (P < 0.05) in spite of the normal T3 levels. This finding may reflect a lasting influence of hyperthyroidism on heart rate or a sensitivity of this group to the anaesthetic used.

3.2. Day of eye opening
Maturation was accelerated in HYPER rats as evidenced by early eye opening (day 10 versus day 12 in CON rats), but was delayed in hypothyroid rats (eyes opened on day 14). Hyper- and hypothyroidism also influenced other developmental parameters: hyperthyroidism facilitated ear and fur growth and led to better neuromuscular coordination, whereas hypothyroidism slowed ear and fur growth and resulted in poorer neuromuscular coordination.

3.3. Functional data
Left and right ventricular functional parameters from 80-day-old rats, 52 days after treatment was stopped, are summarized in Table 1. Right and left ventricular systolic pressure, left ventricular end-diastolic and developed pressure, as well as the maximal rate of pressure development, were all significantly greater in previously HYPER rats (P < 0.05) than in age-matched CON. No lasting effects of neonatal hypothyroidism on cardiac function were detected.


View this table:
[in this window]
[in a new window]

  		Table 1 Left and right ventricular functional data 52 days after treatment stopped in control, hyperthyroid and hypothyroid rats

 
3.4. Body mass and heart mass data
Body mass in HYPER rats was comparable to CON at both 12 and 28 days, but was significantly lower (P < 0.01) 52 days after treatment was discontinued (Table 2). Body mass in HYPO-12 was similar to CON-12 but was significantly lower in HYPO-28, and remained lower 52 days after treatment was stopped (P < 0.01). These results indicate that both neonatal thyroid conditions induced long-term deficits in body growth.


View this table:
[in this window]
[in a new window]

  		Table 2 Basic data for 12- and 28-day-old control, hyperthyroid and hypothyroid rats and for rats 52 days after treatment was stopped

 
Heart mass increased with age in CON and HYPER groups (P < 0.01; Table 2). Compared to age-matched controls, hyperthyroidism induced a sizeable increase in right and left ventricular mass as well as relative heart mass after 12 and 28 days of treatment (P < 0.01). Interestingly, the relative increase in right and left ventricular mass was of similar proportion at 12 and 28 days. It thus appeared that hyperthyroidism equally enhanced the rate of right and left ventricular growth. In HYPO rats, significantly lower right and left ventricular mass (P < 0.05) as well as lower relative heart mass (P < 0.01) were detected after 12 and 28 days of treatment, compared to age-matched controls. A lower right/left ventricular (RV/LV) ratio in HYPO-28 (P < 0.01) indicated that hypothyroidism attenuated the rate of right ventricular growth more than left ventricular growth.

After discontinuing treatment both previously HYPER and HYPO rats demonstrated an increase in right and left ventricular mass over 28-day-old values (P < 0.01). Despite these increases ventricular mass remained lower than CON in previously HYPO rats, whereas only left ventricular mass was lower in previously HYPER rats. These results indicate an inhibitory or attenuating effect of both neonatal thyroid conditions on subsequent left ventricular growth. Relative heart mass was similar among the groups; however, RV/LV ratio was significantly greater in both previously HYPER and HYPO rats (P < 0.01 and P < 0.05, respectively), indicating a larger increase in right than in left ventricular mass after treatment was discontinued.

Percent tissue dry weight was similar among the groups with the exception of HYPER-28, which had a lower percent tissue dry weight than CON-28 (P < 0.05), suggesting the possibility of tissue edema in this group (Table 2).

3.5. Morphometric data
3.5.1. Capillary numerical density
The number of capillaries per unit area decreased with age in CON rats (Fig. GR2); neonatal hyperthyroidism did not affect this developmental pattern. However, no appreciable decrease in capillary numerical density between 12 and 28 days was noted with neonatal hypothyroidism. Greater capillary numerical density (P < 0.05) was observed in HYPER-12 than in CON-12, but this difference was absent in HYPER-28. Considering the induced left ventricular hypertrophy in HYPER-28 rats, these results suggest that hyperthyroidism stimulated marked capillary proliferation, especially in HYPER-12. Capillary numerical density in HYPO-12 and HYPO-28 was not significantly different from age-matched controls, suggesting an attenuation of capillary proliferation in proportion to the lower left ventricular mass. After return to euthyroidism, capillary numerical density was similar among all groups.


Figure 2
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR2 Coronary capillary numerical density (number of capillaries per mm2) in 12- and 28-day-old control (CON), hyperthyroid (HYPER), and hypothyroid (HYPO) rats, and in rats 52 days after treatment was stopped (80 days old). Capillary numerical density decreased significantly with age in HYPER and previously HYPER rats, at P < 0.01, between 28- and 80-day-old CON rats, at P < 0.05, and between 28-day-old HYPO and 80-day-old previously HYPO rats, at P < 0.01. Values are means, error bars = s.e.m. * P < 0.05 vs. CON; {psi} P < 0.01 vs. HYPER.

 
3.5.2. Capillary segment lengths
In CON and HYPO rats proximal capillary segments were always longer than distal segments (P < 0.01; Table 3). No significant difference in proximal and distal segments was found in HYPER-28, suggesting possibly greater longitudinal capillary growth or, alternatively, less branching in distal capillary segments of this group.


View this table:
[in this window]
[in a new window]

  		Table 3 Capillary segment length data for 28-day-old control, hyperthyroid and hypothyroid rats and for rats 52 days after treatment was stopped

 
With age and greater heart mass, capillary segment lengths increased significantly in CON and previously HYPO rats (P < 0.01); proximal capillary segment lengths increased in HYPER (P < 0.01). However, in 80-day-old previously HYPER rats distal capillary segments were not only similar to HYPER-28 distal segment lengths, but were in fact significantly smaller than CON-80 (P < 0.01). These results suggest either an inhibition of subsequent distal segment growth once euthyroidism was re-established, or increased branching in distal portions of the capillary bed.

3.5.3. Total nuclei density
In CON rats, the number of nuclei per unit area decreased between 12 and 28 days of age (P < 0.05) but not thereafter (Fig. GR3). On the other hand, HYPER and HYPO rats showed no initial decrease in nuclei density between 12 and 28 days, but did show a decrease between 28 and 80 days (P < 0.05). At 12 days of age, HYPO rats had a lower nuclei density compared to age-matched CON (P < 0.05).


Figure 3
View larger version (25K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR3 Total nuclei density (number of nuclei per mm2) in 12- and 28-day-old control (CON), hyperthyroid (HYPER), and hypothyroid (HYPO) rats, and in rats 52 days after treatment was stopped (80 days old). Total nuclei density decreased significantly with age, at P < 0.05, between 12- and 28-day-old CON, and between 28-day-old HYPER and HYPO rats and 80-day previously HYPER and HYPO rats. Values are means, error bars = s.e.m. * P < 0.05 vs. CON; {delta} P < 0.05 vs. 28-day-old.

 
3.5.4. Relative nuclei proportions
The relative proportions of endothelial (ENDO), myocyte (MYO), and ‘other’ (OTHER) nuclei (Table 4) changed between 12 and 28 days in both CON and HYPER rats (P < 0.05). This redistribution was delayed in HYPO rats. Both neonatal hypo- and hyperthyroidism induced significant changes in nuclei proportions at 12 and 28 days of age as compared to CON (P < 0.01). Fifty-two days after treatment was discontinued nuclei proportions were comparable among the groups, suggesting no lasting effect of either neonatal thyroid condition.


View this table:
[in this window]
[in a new window]

  		Table 4 Proportion of endothelial, myocyte, and ‘other’ nuclei in 12- and 28-day-old control, hyperthyroid and hypothyroid rats and in rats 52 days after treatment was stopped

 
3.5.5. PCNA-positive nuclei
An increase in the number of PCNA-positive endothelial (ENDO +) nuclei between CON-12 and CON-28 indicated endothelial cell proliferation, likely reflecting the occurrence of capillary growth (Fig. GR4). The highest percentage of ENDO + nuclei occurred at 12 days in HYPER rats but at 28 days in control rats. By day 28, the higher percentage of ENDO + nuclei in HYPER rats had decreased to a similar level as CON-28.


Figure 4
View larger version (18K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR4 Number of PCNA-positive endothelial (ENDO), myocyte (MYO), and ‘other’ (OTHER) nuclei in 12- and 28-day-old control (CON), hyperthyroid (HYPER), and hypothyroid (HYPO) rats, and in rats 52 days after treatment was stopped (80 days old). Values are means, error bars = s.e.m. * P < 0.05 vs. CON; {delta} P < 0.05 vs. HYPER.

 
The number of PCNA-positive myocyte nuclei (MYO +) decreased with age in all groups. No statistical difference in MYO + nuclei among the groups was detected at any age. The lack of MYO + nuclei in 80-day-old control rats agrees with the idea that myocyte proliferation is rare in adult animals; this phenomenon was not changed by either neonatal hyper- or hypothyroidism.

The occurrence of PCNA-positive ‘other’ (OTHER +) nuclei decreased with age in CON and HYPER, and between HYPO-12 and HYPO-28. Fewer OTHER + nuclei were observed in HYPO-12 and HYPO-28 than in age-matched CON, though neither reached statistical significance. Interestingly, for each nucleus type (i.e., ENDO, MYO, OTHER), HYPER-12 had more PCNA-positive nuclei than CON-12, suggesting that neonatal hyperthyroidism enhanced the proliferation of all cell types in myocardial tissue.


   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Cardiac hypertrophy and accelerated maturation were produced in neonatal rats treated with excess triiodothyronine (T3), whereas attenuated cardiac growth and delayed maturation were induced in neonatal rats treated for a similar period with the anti-thyroid goitrogen, propylthiouracil. These results agree with other neonatal studies[10, 11, 13]. Absolute changes in body and heart mass, which were noted in the present study, differ from those observed in a similar neonatal study that used the same treatment protocol [24]. These discrepancies likely resulted from the fact that rats used in the present study were different from those in the other study. A similar explanation may be advanced for the discrepancy in serum T3 levels between the present and other study. Additionally, seasonal effects or diurnal variations in plasma TSH levels[25] cannot be excluded as alternative explanations.

In the present study, the proportional increase in ventricular mass was different from the predominant right ventricular hypertrophy reported in other neonatal [11] and adult studies [2]. It is possible that treatment every second day may not produce the same degree of increased hemodynamic load as daily treatment, and this in part may account for the discrepancy. The predominant attenuation of right ventricular growth relative to left in hypothyroid rats also differs from reports of proportionally lower ventricular mass in neonatal [11] and adult hypothyroid rats[2].

Notably, the growth mechanisms governing the elicited cardiac growth response differ between neonatal and adult rats. For instance, in adult hyperthyroid rats myocyte hypertrophy is believed to be the primary mechanism for increased cardiac mass [26], whereas in neonatal hyperthyroid rats both myocyte hyperplasia and myocyte hypertrophy play a role, though to varying degrees [12]. Comparison of PCNA labelling of myocytes in neonatal hyperthyroid rats from this study with PCNA labelling of myocytes in adult hyperthyroid rats [20] provides support for the above hypothesis. PCNA labelling was absent in myocytes from adult hyperthyroid rats compared to the larger number of PCNA immunoreactive myocytes found in hearts from neonatal hyperthyroid rats.

Conversely, adult hypothyroid rats exhibit signs of myocyte atrophy [27], whereas neonatal hypothyroid rats experience attenuation of myocyte growth and proliferation[28]. The smaller number of PCNA-positive myocyte nuclei in neonatal hypothyroid rats from this study and the absence of PCNA labelling in myocytes from adult hypothyroid rats [20] is in agreement with this idea. The reduction in cell proliferation with early neonatal hypothyroidism has been previously described for other body organs [28] and is supported in the present study by the smaller number of PCNA-positive nuclei and similarity in overall nuclei density between 12- and 28-day-old hypothyroid rats. Thus, alterations in thyroid status during the neonatal period not only accelerate or delay maturation, but similarly influence the degree of cell proliferation and cardiac growth.

4.1. Coronary capillary morphometry
Commensurate with the enhanced cardiac growth induced by hyperthyroidism, enhanced capillary proliferation was also observed in these rats. After 12 days of T3 treatment, the significantly larger capillary numerical density in spite of increased cardiac mass, and significantly greater number of PCNA-positive endothelial nuclei, provides evidence indicative of considerable capillary proliferation. After 28 days of T3 treatment, capillary proliferation declined to a level which was proportional to the increase in left ventricular mass. The decrease in number of PCNA-positive endothelial nuclei supports this finding. The induction of capillary proliferation in response to hyperthyroidism has previously been observed in adult[1–3] and senescent hyperthyroid rats [4] and in adult hyperthyroid pigs [29]. In a comparable study, hyperthyroidism was found to stimulate coronary arteriolar growth in neonatal rats [24]. Increased mechanical stress on endothelial walls has previously been proposed as a stimulus for capillary growth [30], and in the case of hyperthyroidism would likely result from greater coronary blood flow [31] and blood viscosity [32].

Interestingly, the similarity in length between proximal and distal capillary segments in 28-day-old hyperthyroid rats suggests that growth of segments along the longitudinal axis (i.e., capillary elongation) was more pronounced in distal regions of the capillary bed than in proximal. One possible explanation for this finding may be based on the increased metabolism and myocardial oxygen consumption produced by hyperthyroidism [31]. This increase may exacerbated the low pO2 found in distal portions of the capillary bed [33], which in turn may stimulate distal capillary growth.

In contrast, neonatal hypothyroidism slowed the formation of new capillary branches or connections in proportion to the reduction in cardiac growth, thus maintaining capillary numerical density. However, data from capillary segment length analysis indicated that even though formation of new capillaries was attenuated, increases in the length of existing capillaries were not appreciably affected. The similarity in number of PCNA-positive endothelial nuclei between 12-day-old control and hypothyroid rats indicates some endothelial cell proliferation, supporting the occurrence of capillary elongation. These findings are different from the increased capillary numerical density previously reported in adult hypothyroid rats [2] and rabbits [34], and in neonatal hypothyroid rats [24]. The present findings are comparable, nonetheless, to the attenuated arteriolar growth which was noted in neonatal hypothyroid rats [24].

The discrepancy in capillary morphometry between this and the other neonatal study may be due to the different histological methods used in each study. An alternative explanation may be that the effect of hypothyroidism on capillary growth in neonatal rats is variable. For instance, lack of thyroid hormone markedly effects cells that are in the proliferative phase of growth. As significant capillary growth occurs during the neonatal period, hypothyroidism would slow proliferation of endothelial cells. However, increased mechanical stress resulting from induced brady-cardia may simultaneously be providing a stimulus for capillary proliferation. The overall result of these opposing influences on capillary growth may lead to varied response. In the adult rat, capillary growth is negligible and thus only increased mechanical stress on the endothelium would be acting to provide a stimulus for capillary proliferation.

Based on the results of this study, one may speculate that the decline in capillary proliferation that occurred during postnatal development depends more on maturation than on the rate of cardiac growth. The rate of cardiac growth may be considered from an increase in heart mass over time. Accelerated maturation resulting from hyperthyroidism has been previously reported (e.g., [35]) and was indicated in the present study (i.e., early eye opening). In both control and hyperthyroid rats the rate of cardiac growth was similar between 12 and 28 days of age (an increase in LV mass of ~ 142% in both groups), yet maturation was accelerated in hyperthyroid rats. During this time, the decline in capillary numerical density was greater in hyperthyroid (decrease of ~ 34%) than control rats (decrease of ~ 17%). Thus, the decline in capillary proliferation which results in decreased capillary numerical density between 12 and 28 days appears to be consistent with a dependence on tissue maturation rather than the rate of cardiac growth. Additional support for this postulate is the finding that the number of PCNA-labeled endothelial cells decreased earlier in hyperthyroid rats than in control.

4.2. Long-term effects of altered neonatal thyroid status
Both neonatal hyper- and hypothyroidism resulted in long-term body and cardiac growth deficits, albeit likely by different mechanisms. With respect to cardiac growth, it has been proposed that neonatal hyperthyroidism leads to a reduction in myocyte numbers by hastening the transition from hyperplastic to hypertrophic growth [12]. The smaller number of myocytes in these hearts are not able to produce the same increase in cardiac mass with age, as in normal hearts. On the other hand, in neonatal hypothyroidism, though the number of cardiac myocytes may not be affected due to the delay in transition from hyperplastic to hypertrophic growth, reduction in other hormones (e.g., growth hormone) which play an important role in cardiac development may compromise cardiac growth subsequent to re-establishment of euthyroidism [10].

No long-term changes in capillarization were observed with neonatal hypothyroidism. The similarity of distal segment lengths between 28-day-old hyperthyroid and 80-day-old previously hyperthyroid rats, together with increasing left ventricular mass during this period, may be explained by two possible, yet different scenarios. The first scenario would be that after treatment was stopped, formation of additional connecting capillary branches occurred primarily in distal portions of the capillary bed. The second would be that subsequent growth of capillaries in distal portions of the capillary bed was inhibited by neonatal hyperthyroidism. Since formation of new capillaries would likely produce an increase in capillary numerical density and this was not observed, the second explanation seems more plausible. In fact, in a similar study, neonatal hyperthyroidism was found to result in an absence of subsequent arteriolar growth after return to euthyroidism[24]. The absence of arteriolar, and likely capillary, growth after stopping T3 treatment may be due to removal or reduction of the growth stimulus (e.g., hypoxia, direct effect of T3, mechanical stimulus, etc.) that was present with treatment.

Furthermore, neonatal hyperthyroidism appeared to induce lasting positive effects on heart rate and on indices of systolic and diastolic mechanical function of the heart. The explanation of these observations is unclear and many possible mechanisms may be taken into account. Thyroid hormones are known to induce a variety of direct effects on cardiac muscle as well as influence heart function indirectly (e.g., via effects on peripheral circulation or interaction with the adrenergic nervous system) [36]. However, these changes have been shown to return to normal after stopping treatment, at least in adult animals [37]. As in adults, excess thyroid hormone during the neonatal period induces positive chronotropic and inotropic effects on the heart that are associated, for example, with altered calcium transport by cardiac membranes [11, 35] and changes at the level of the contractile apparatus [38, 39]. It is unknown whether these changes are reversible as those induced during adulthood. One may even speculate that at the time these rats were examined (i.e., 52 days after stopping treatment), some of the changes induced during the neonatal period may not have completely returned to normal. The persistent lower number of cardiac myocytes resulting from neonatal hyperthyroidism [12] may through some undetermined mechanism produce the inotropic and chronotropic effects observed. It has been suggested that the presence or absence of myocyte proliferation during adaptive heart growth in response to pressure overload can impact on the maintenance of normal myocardial function[40]. An alternative explanation for these findings may be an increased sensitivity of these rats to the anaesthetic used. To the best of our knowledge, however, no other studies in hyperthyroid rats have reported any anomalous responses to anaesthesia.

Results from the present study indicate that neonatal hyperthyroidism increases cardiac growth and induces substantial capillary proliferation, whereas attenuation of cardiac growth with reduced capillary proliferation results from neonatal hypothyroidism. Both neonatal thyroid conditions produced long-term deficits in body and cardiac growth, yet only neonatal hyperthyroidism induced a possible inhibition on subsequent growth of capillaries in distal portions of the capillary bed. Additionally, neonatal hyperthyroidism produced a lasting positive influence on the inotropic and chronotropic characteristics of left ventricular function, though a plausible explanation or physiological relevance has yet to be determined. One other interesting finding, based on our results, is that the pattern of capillary growth during the neonatal period of development (i.e., rapid initial capillary proliferation with subsequent decline shortly after weaning) may depend more on tissue maturation than on the rate of cardiac growth.


   Acknowledgements
 
The authors sincerely thank Mrs. Ching Kuo and Mrs. Barbara Hebert for all of their valuable assistance. This work was supported by a grant from the Heart and Stroke Foundation of Canada (KR) and a Research Traineeship from the Heart and Stroke Foundation of Canada (MIH).


   Notes
 
* Corresponding author. Tel. + 1 613 562-5800, ext. 8384; Fax + 1 613 562-5434; E-mail: krakusan@labsun1.med.uottawa.ca Back


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Wachtlova M, Ostadal B, Mares V. Thyroxine-induced cardiomegaly in rats of different age. Physiol Bohemoslov (1985) 34:385–394.[Medline]
  2. Heron MI, Rakusan K. Geometry of coronary capillaries in hyperthyroid and hypothyroid rat heart. Am J Physiol (1994) 267:H1024–H1031.[Web of Science][Medline]
  3. Chilian WM, Wangler RD, Peters KG, Tomanek RJ, Marcus ML. Thyroxine-induced left ventricular hypertrophy in the rat. Anatomical and physiological evidence for angiogenesis. Circ Res (1985) 57:591–598.[Abstract/Free Full Text]
  4. Tomanek RJ, Connell PM, Butters CA, Torry RJ. Compensated coronary microvascular growth in senescent rats with thyroxine-induced cardiac hypertrophy. Am J Physiol (1995) 268:H419–H425.[Web of Science][Medline]
  5. Rakusan K, Jelinek J, Korecky B, Soukupova M, Poupa O. Postnatal development of muscle fibres and capillaries in the rat heart. Physiol Bohemoslov (1965) 14:32–37.[Medline]
  6. Batra S, Rakusan K. Capillary network geometry during postnatal growth in rat hearts. Am J Physiol (1992) 262:H635–H640.[Web of Science][Medline]
  7. Rakusan K, Flanagan MF, Geva T, Southern J, Van Praagh R. Morphometry of human coronary capillaries during normal growth and the effect of age in left ventricular pressure-overload hypertrophy. Circulation (1992) 86:38–46.[Abstract/Free Full Text]
  8. Flanagan MF, Aoyagi T, Currier JJ, Colan SD, Fujii AM. Effect of young age on coronary adaptations to left ventricular pressure overload hypertrophy in sheep. J Am Coll Cardiol (1994) 24(7):1786–1796.[Abstract]
  9. Hedge GA, Colby HD, Goodman RL. Hormonal control of growth. In: Clinical Endocrine Physiology. Toronto: WB Saunders Company (1987) 317–339.
  10. Meisami E. Complete recovery of growth deficits after reversal of PTU-induced postnatal hypothyroidism in the female rat: a model for catch-up growth. Life Sci (1984) 34:1487–1496.[CrossRef][Web of Science][Medline]
  11. Kolar F, Seppet EK, Vetter R, et al. Thyroid control of contractile function and calcium handling in neonatal rat heart. Eur J Physiol (1992) 421:26–31.[CrossRef][Web of Science][Medline]
  12. Gerdes AM, Kriseman J, Bishop SP. Changes in myocardial cell size and number during the development and reversal of hyperthyroidism in neonatal rats. Lab Invest (1983) 48(5):598–602.[Web of Science][Medline]
  13. Moussavi R, Meisami E, Timiras PS. Compensatory cell proliferation and growth in the rat heart after postnatal hypothyroidism. Am J Physiol (1985) 248:E381–E387.[Web of Science][Medline]
  14. Dubeck JK, Penney DG, Brown TR, Sharma P. Thyroxine treatment of neonatal rats suppresses normal and stress-stimulated heart cell hyperplasia and abolishes persistent cardiomegaly. J Appl Cardiol (1989) 4:195–205.[Web of Science]
  15. Rakusan K, Turek Z. Protamine inhibits capillary formation in growing rat hearts. Circ Res (1985) 57:393–398.[Abstract/Free Full Text]
  16. Mirsky I, Pasipoularides A. Clinical assessment of diastolic function. Prog Cardiovasc Dis (1990) 32:291–318.[CrossRef][Web of Science][Medline]
  17. Hansen-Smith FM, Watson L, Lu DY, Goldstein I. Griffonia simplicifolia I: fluorescent tracer for microcirculatory vessels in nonperfused thin muscles and sectioned muscle. Microvasc Res (1988) 36:199–215.[CrossRef][Web of Science][Medline]
  18. Christie KN, Thomson C. Bandeiraea simplicifolia I lectin demonstrates significantly more capillaries in rat skeletal muscle than enzyme methods. J Histochem Cytochem (1989) 37:1303–1304.[Free Full Text]
  19. Hansen-Smith FM, Blackwell L. Acute changes in muscle capillarity of hypoxic mice. FASEB J (1989) 3:A270.
  20. Heron MI, Rakusan K. Proliferating cell nuclear antigen (PCNA) detection of cellular proliferation in hypothyroid and hyperthyroid rat hearts. J Mol Cell Cardiol (1995) 27(7):1393–1403.[CrossRef][Web of Science][Medline]
  21. Lojda Z. Studies on dipeptidyl (amino) peptidase IV (glycl-proline napthylamidase). Histochemistry (1979) 59:153–166.[CrossRef][Web of Science][Medline]
  22. Batra S, Kuo C, Rakusan K. Spatial distribution of coronary capillaries: A-V segment staggering. In: Rakusan K, Biro G, Goldstick TK, Turek Z, eds. Oxygen Transport to Tissue XI. New York: Plenum Press, 1989: 241–247.
  23. Rakusan K, Batra S, Heron MI. A new approach for quantitative evaluation of coronary capillaries in longitudinal sections. In: Hogan MC, Mathieu-Costello O, Poole DC, Wagner PD, eds. Oxygen Transport to Tissue XVI. New York: Plenum Press, 1994:407–415.
  24. Heron MI, Rakusan K. Short and long-term effects of neonatal hypo- and hyperthyroidism on coronary arterioles in rat. Am J Physiol 1996 (in press).
  25. Jolin T, Pallardo LF, Pericas I. Diurnal variations of plasma and pituitary thyrotrophin in adult male and female rats. Acta Endocrinol (1976) 83:745–751.[Medline]
  26. Campbell S, Korecky B, Rakusan K. Remodelling of myocyte dimensions in hypertrophic and atrophic rat hearts. Circ Res (1991) 68:984–996.[Abstract/Free Full Text]
  27. Liu Z, Gerdes AM. Influence of hypothyroidism and the reversal of hypothyroidism on hemodynamics and cell size in the adult rat heart. J Mol Cell Cardiol (1990) 22:1339–1348.[CrossRef][Web of Science][Medline]
  28. Brasel J, Winick M. Differential cellular growth in the organs of hypothyroid rats. Growth (1970) 34:197–207.[Web of Science][Medline]
  29. Breisch EA, White FC, Hammond HK, Flynn S, Bloor CM. Myocardial characteristics of thyroxine stimulated hypertrophy. A structural and functional study. Basic Res Cardiol (1989) 84:345–358.[Web of Science][Medline]
  30. Hudlicka O. Capillary growth: role of mechanical factors. News Physiol Sci (1988) 3:117–120.[Abstract/Free Full Text]
  31. Talafih K, Grover GJ, Weiss HR. Effect of T4-induced cardiac hypertrophy on O2 supply-consumption balance during normoxia and hypoxia. Am J Physiol (1984) 246:H374–H379.[Web of Science][Medline]
  32. Kossler A, Hagmuller K, Winkler R. The effects of experimental hypo- and hyperthyroidism on blood viscosity and other blood parameters in the rat. Biorheology (1987) 24:769–774.[Web of Science][Medline]
  33. Vetterlein F, Prange M, Lubrich D, Pedina J, Neckel M, Schmidt G. Capillary perfusion pattern and microvascular geometry in heterogeneous hypoxic areas of hypoperfused rat myocardium. Am J Physiol (1995) 268:H2183–H2194.[Web of Science][Medline]
  34. Tomanek RJ, Barlow PA, Connell PM, Chen Y, Torry RJ. Effects of hypothyroidism and hypertension on myocardial perfusion and vascularity in rabbits. Am J Physiol (1984) 265:H1638–H1644.
  35. Wibo M, Kolar F, Zheng L, Godfraind T. Influence of thyroid status on postnatal maturation of calcium channels, β-adrenoceptors and cation transport ATPases in rat ventricular tissue. J Mol Cell Cardiol (1995) 27:1731–1743.[CrossRef][Web of Science][Medline]
  36. Polikar R, Burger AG, Scherrer U, Nicod P. The thyroid and the heart. Circulation (1993) 87:1435–1441.[Abstract/Free Full Text]
  37. Cihak R, Kolar F, Pelouch V, Prochazka J, Ostadal B, Widimsky J. Functional changes in the right and left ventricle during the development of cardiac hypertrophy and after its regression. Cardiovasc Res (1992) 26:845–850.[Abstract/Free Full Text]
  38. Chizzonite RA, Zak R. Regulation of myosin isozyme composition in fetal and neonatal rat ventricle by endogenous thyroid hormones. J Biol Chem (1984) 259:12628–12632.[Abstract/Free Full Text]
  39. Dieckman LJ, Solaro RJ. Effect of thyroid status on thin-filament Ca2+ regulation and expression of troponin I in perinatal and adult rat hearts. Circ Res (1990) 67:344–351.[Abstract/Free Full Text]
  40. Dowell RT, McManus RE. Pressure-induced cardiac enlargement in neonatal and adult rats: left ventricular functional characteristics and evidence of cardiac muscle cell proliferation in the neonate. Circ Res (1978) 42(3):303–310.[Free Full Text]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 EndocrinologyHome page
A. Makino, J. Suarez, H. Wang, D. D. Belke, B. T. Scott, and W. H. Dillmann
Thyroid Hormone Receptor-{beta} Is Associated with Coronary Angiogenesis during Pathological Cardiac Hypertrophy
Endocrinology, April 1, 2009; 150(4): 2008 - 2015.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Heron, M. I.
Right arrow Articles by Rakusan, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Heron, M. I.
Right arrow Articles by Rakusan, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?