We thank Dr Libonati for his comment to our recent publication concerning the effect of exercise on spontaneously hypertensive rats with established hypertension (da Costa Rebelo et al. 2012). This is a very interesting field with some conflicting data, as summarized by Libonati in his Letter to the Editor. The author raised the question of whether rats that were removed from the study due to severe failure are indeed developing heart failure. Specifically, he requested data on physical activity at the time when rats were removed due to signs of failure. We agree with his comment and consider such information to be of interest for the readership. As outlined in our study we removed three rats from the spontaneously hypertensive group. Another rat died spontaneously. Table 1 gives the running performance of these three rats within the last 5 weeks:

Table 1. Table 1
Weeks remaining 43210
Running distance (km per week)Rat 137.819.619.018.715.6
 Rat 260.364.032.725.715.7
 Rat 319.310.312.811.87.5
Running time (h:min per week)Rat 119:2211:4811:5711:159:44
 Rat 220:0925:0814:1112:048:19
 Rat 39:546:086:236:484:39
Running velocity (km h-1)Rat 11.1911.661.591.661.60
 Rat 22.482.552.312.122.91
 Rat 31.951.692.011.731.62

All three rats responded positively to the training protocol, as shown by a reduction in resting heart rate within 4 weeks upon starting (rat 1: 451–408 beats min-1; rat 2: 432–376 beats min-1; rat 3: 473–388 beats min-1). Lung wet weights for the three rats were 547, 544 and 951 mg (mm tibia length)-1, respectively. The mean lung wet weight of sedentary SHR animals was 481 ± 23 mg mm-1. We hope these data clarify the severe failure of rats removed from the study. Subsequent functional and molecular analysis confirmed the existence of decompensated heart failure.

Another issue raised by Dr Libonati is on our measurement of left ventricular developed pressure in the Langendorff analysis. The data in the original publication were normalized to heart weight. The non-normalized values for left ventricular developed pressure were 141 mmHg for SHR-trained and 177 mmHg for SHR-sedentary animals. These values are in agreement with all the literature cited in Libonati's letter. We agree that additional echocardiography would have been of interest but this was not possible at the time of our study.

Dr Libonati also raised the question of the difference between free running wheel exercise (short running bouts) and endurance training (low-intensity treadmill exercise). We have previously discussed this issue in greater detail. One remaining issue is that the rat per se is not a long-distance runner as it cannot easily control its body temperature. Therefore, it is at present unclear what treadmill protocols really induce in these animals. Nevertheless, we completely agree to his comment that a more active scientific dialogue is required to understand the interaction between hypertension, ageing and physical activity.

In another response to our recent publication, Dr Ibrahim and colleagues asked, ‘what do we know and how can we learn more about exercise, load and remodelling?’ We thank them for their interesting comment. The authors discuss the importance of a subsequent study that analyses the effect of voluntary exercise on spontaneous hypertension at earlier, probably pre-hypertensive, time points and address whether under such conditions the ability of the rats to adapt to the exercise load with respect to cardiac remodelling is different. In other words, does exercise modify the development of a hypertension-induced hypertrophy in a way that is different from how it affects an existing hypertrophic heart. The results of this study are due to be published in the near future. However, we can assume that it will probably not be straightforward to find a strict correlation between the duration of hypertension and the loss of adaptation to the increased workload. In a recent meta-analysis on published studies with spontaneously hypertensive rats (18 reports; 410 rats) we have shown that an effect of blood pressure lowering was reproducibly found for young or even pre-hypertensive rats (Schlüter et al. 2010). Interestingly, this effect disappears with time. Therefore, the effect of exercise on blood pressure in spontaneously hypertensive rats is most remarkable in studies with a low level of exercise, the use of young or even pre-hypertensive rats, and a short follow-up period. It must therefore be classified as a transient phenomenon that describes an early adaptation to an increased level of performance.

Although these studies seem to be conclusive, they ignore a main problem in the field. How can we define a physiological model of exercise in rats? Rats are nocturnal animals that normally run using short running bouts of 1–2 min at relative high velocity. This is different from humans, which are more adapted to long-lasting running at low velocity. Such protocols can be adapted easily to rats in terms of treadmill protocols in which rats normally run to a much lower level with regard to distance and velocity but over a longer time (usually 30–60 min day-1). This raises two questions. First, can we define such a model as a physiological running profile of a rat and if not what does this tell us about physiological adaptations of the heart to this stress? Secondly, how does a low level of exercise add something in comparison with so-called sedentary rats? In fact, even sedentary rats move considerable distances in their cages during the night and it is difficult to define the additional training effect involved. Treadmill protocols are normally adaptations of rats to a non-physiological moving performance. This can easily be seen by the fact that they have to be treated with electrodes before proceeding with the protocol. Therefore, it is quite difficult to compare exercise-dependent effects between rodent models and humans. However, it should be clearly noted that in our study the extent of exercise was not sufficient to induce a physiological type of hypertrophy in non-hypertensive control rats. Thus, even the relatively high level of voluntary exercise performed by these rats does not exert any harm to the rats per se. This might indicate that it is not simply the loss of adaptation to high physical activity that induced the mal-adaptive hypertrophy in our rats.

Finally, we note that clinical studies showing an effect of exercise on hypertension always involve lifestyle changes rather than high physical activity alone. In such studies there are additional effects, such as reduced body weight, that will per se contribute to an effect of blood pressure lowering and psychological effects.

In conclusion, in answer to the question raised by Dr Ibrahim and colleagues: No, we do not yet know everything about exercise, load and remodelling. But let us go one step further and determine the complex interaction between these three important determinants of heart function.


  1. Top of page
  2. References