Article Text

Download PDFPDF

Reduced 24 hour ambulatory blood pressure and abnormal heart rate variability in patients with dysorexia nervosa
  1. N D Cong1,
  2. T Saikawa2,
  3. R Ogawa3,
  4. M Hara3,
  5. N Takahashi3,
  6. T Sakata3
  1. 1Department of Medicine, Hanoi, Vietnam
  2. 2Department of Cardiovascular Science, Division of Laboratory Science, School of Medicine, Oita Medical University, Hasama, Oita, Japan
  3. 3Department of Medicine, Internal Medicine I, School of Medicine, Oita Medical University
  1. Correspondence to:
    Professor Tetsunori Saikawa
    Department of Cardiovascular Science, School of Medicine, Oita Medical University, Hasama, Oita, 879-5593, Japan;

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Dysorexia nervosa (DN) includes two types of eating disorder: anorexia nervosa (AN) and bulimia nervosa (BN). AN is a syndrome characterised by changes in eating habits, distortions of body shape, and a refusal to maintain even minimally normal body weight. BN is a syndrome encompassing repeated episodes of binge eating, followed by inappropriate compensatory behaviours such as self induced vomiting, misuse of laxatives and diuretics, fasting or excessive exercise. Both AN and BN predominantly affect young females, in whom the prevalence is approximately 10 times higher than in males, and both are important social health concerns in developed countries.1

The following study was designed to explore the underlying autonomic dysfunction in DN; six patients with AN, eight with BN and 11 healthy subjects were assessed. The circadian changes in blood pressure (BP), heart rate (HR), and heart rate variability (HRV) of the patients were measured.


Twenty four hour ambulatory BP and HR were recorded using a portable multibiomedical recorder (TM-2425, A&D Co, Tokyo, Japan). The measurements of BP and HR were repeated every 30 minutes in the waking period from 6 am to 11 pm and every 60 minutes during sleep from 11 pm to 6 am. Systolic and diastolic BP were measured by both Korotkoff and oscillometric methods. Frequency domain indices of HRV were simultaneously calculated from the data. Ranges of 0.05–0.15 Hz and 0.15–0.40 Hz were defined as low frequency (LF) and high frequency (HF) component ranges, respectively. The power HF component was assumed as an index of parasympathetic activity, and the ratio of the powers of LF and HF (LF/HF) as an index of sympathetic activity.2


There was no significant difference in age, height or body weight between the BN and control groups (AN 33.6 kg, BN 47.6 kg, control 50.9 kg, respectively), or between the body mass indices (BMI) of BN and control groups (AN 13.9 kg/m2, BN 19.1 kg/m2, 20.3 kg/m2, respectively). Both body weight and BMI in the AN group were lower than in the BN and control groups. As shown in table 1, AN and BN patients exhibited lower mean 24 hour systolic BP than controls. The mean 24 hour diastolic BP was lower than the controls in the AN group but not in the BN group. The mean 24 hour HR revealed no significant difference among AN, BN, and control groups (table 1). Both systolic and diastolic BP during wakefulness were lower in AN and BN groups than in controls (table 1). The incidence of the non-dipper BP pattern in the AN and BN groups was 66.7% and 62.5%, respectively, higher than in the control group (9.1%) (p  =  0.0276 and 0.0408, respectively).

Table 1

Mean systolic and diastolic blood pressure and mean heart rate during the whole day, wakefulness and sleep in anorexia nervosa (AN), bulimia nervosa (BN), and control groups (CN)

The power of HF in AN and BN patients was increased significantly by 10.3% and 12.5%, respectively, compared to controls (p  =  0.0462 and 0.0163, respectively). This in turn resulted in a significantly lower LF/HF ratio in the 24 hour analysis in AN and BN groups compared to controls. The HF power in AN and BN groups remained higher than controls during waking (p  =  0.0492 and 0.0091, respectively) and during sleep (p  =  0.0244 and 0.0102, respectively). Consequently, the LF/HF ratio in AN and BN groups was concomitantly lower than in controls during both waking and sleeping, respectively.


Analysis of BP and HR in AN patients is confounded by the extremely low body weight of the subjects which, as shown in the present study, inevitably induces distinct neuroendocrinological modulation of BP and HR via mechanisms such as altered adrenal and thyroid function. Thus, it is possible that the observed changes in BP and HR do not reflect the underlying mechanism(s) of AN but rather are secondary to such neuroendocrinological modulation. To investigate this, we simultaneously analysed BN patients and found that patients with both AN and BN exhibited similar modulation of BP and HR regardless of the differences in body weight and BMI between the two conditions.

The 24 hour ambulatory BP was reduced in both AN and BN groups and revealed a higher incidence of the non-dipper BP pattern in BN (62.5%) and AN (66.7%) than in controls (9.1%). No significant differences in the incidence of the non-dipper BP pattern or in indices reflecting autonomic nerve activity were noted between AN and BN groups, although the BMI was significantly different between the two. In addition, both AN and BN groups were associated with increased HF power and a decreased LF/HF ratio throughout the 24 hour period of wakefulness and sleep. The results were consistent with lower plasma concentrations of noradrenaline (norepinephrine) and lower BP in AN and BN groups than in controls3,4 together with a rapid response to the infusion of isoproterenol.3 Thus, the low LF/HF and higher incidence of non-dipper BP pattern in AN and BN suggest reduced activity of the sympathetic nervous system in these patients.5

In conclusion, we have shown the presence of abnormalities in BP patterns in both AN and BN patients characterised by increased incidence of the non-dipper BP pattern, low BP during wakefulness, and disturbances of the autonomic nervous system with increased parasympathetic and decreased sympathetic activity throughout the 24 hour period. The results suggest the dysfunction of the hypothalamus in DN patients may play an important contributory role to this abnormal BP dynamic pattern and enhance our understanding of autonomic cardiac regulation in AN and BN.