| |
Autonomic dysfunction in HIV patients on antiretroviral therapy: studies of heart rate variability
|
| |
| |
Download the PDF here
Clinical Physiology and Functional Imaging November 2007
CROI/2013: Fall Frequency and Associated Factors among Men and Women with or at Risk for HIV Infection - (04/05/13)....
from jules: this study reported at CROI this year, for the first time a study reports in HIV+ individuals "clinical symptoms of floating, lightheadedness, faintness, and off-balance were more prevalent in HIV+ persons and were associated with fall frequency....if persistent these symptoms may increase the risk for falls with aging." Although the study did not make any references to autonomic dysfunction these are symptoms associated with autonomic dysfunction in the general population.
"Autonomic dysfunction has been shown to be common in patients with HIV and AIDS....Most studies have been performed in untreated patients or patients with very advanced disease.......Most previous studies of autonomic function in HIV patients have used simple autonomic function tests. However, the use of heart rate variability (HRV) with spectral analysis allows for the assessment of relative contribution of sympathetic and parasympathetic involvement........The aim of the present study was therefore to study the presence of autonomic dysfunction in a population of optimally treated HIV positives receiving ART. To do so, we performed analysis of HRV to assess the relative contribution of the sympathetic and parasympathetic nervous system."
-----------------------------
Dysautonomia (or autonomic dysfunction) is any disease or malfunction of the autonomic nervous system (ANS). The autonomic nervous system controls a number of functions in the body, such as heart rate, blood pressure, digestive tract peristalsis, and sweating, amongst others. Dysfunction of the ANS can involve any of these functions......mitochondrial diseases can be a cause......The outlook for patients with dysautonomia depends on the particular diagnostic category. Some forms of dysautonomia resolve over time and are not life threatening, even if life-changing in the form of minor to major limitations in activities of daily living. Patients with chronic, progressive, generalized dysautonomia in the setting of central nervous system degeneration have a generally poor long-term prognosis. Death can occur from pneumonia, acute respiratory failure, or sudden cardiopulmonary arrest in such patients
http://en.wikipedia.org/wiki/Dysautonomia
The symptoms of dysautonomia are numerous and vary widely from person to person. Since dysautonomia is a full-body condition, a large number of symptoms may be present that can greatly alter a person's quality of life. Each patient with dysautonomia is different-some are affected only mildly, while others are left completely bed-ridden and disabled.
The primary symptoms present in patients with dysautonomia are:
· Excessive fatigue
· Excessive thirst (polydipsia)
· Lightheadedness, dizziness or vertigo
· Feelings of anxiety or panic (not mentally induced[citation needed])
· Rapid heart rate or slow heart rate
· Orthostatic hypotension, sometimes resulting in syncope[1] (fainting)
------------------------------
AUTHORS CONCLUDE: "HIV patients in long-standing HAART have increased resting heart rate and decreased short-term HRV indicating parasympathetic dysfunction......"
"Development of parasympathetic autonomic dysfunction may be especially dangerous if patients at the same time develop ischaemic heart disease as the combination may be arrhythmogenic.....we found a clear and selective parasympathetic dysfunction with no indication of sympathetic dysfunction. Compared with studies of advanced disease in untreated patients where in general both sympathetic and parasympathetic dysfunction were present, it may be speculated whether the dysfunction observed in our population has a different pathogenesis than the one observed in the pretreatment era. The autonomic dysfunction in untreated patients with advanced disease is generally believed to be caused by HIV-1 virus itself, which is well known to be neurotropic. In contrast, the selective parasympathetic dysfunction found in our patients could rather be caused by the anti-retroviral drugs used that are known to be neurotoxic. Accordingly, some of the nucleoside reverse transcriptase inhibitors especially stavudine are known to affect mitochondrial function and may lead to depletion of the neuronal mitochondrial DNA and to an inhibition of DNA polymerase (Dalakas et al., 2001; Peltier & Russell, 2002; Authier & Gheradi, 2003). These effects on the mitochondria may be related to the polyneuropathy that develops in some of the patients receiving these drugs. Although a total of eight of our patients had received stavudine previously, only one patient was on the treatment when the examination was performed. As all the patients did have a lactate in the normal range, we do not believe that stavudine induced hyperlactataemia is the reason for our findings. The diabetogenic effect of ART could lead to parasympathetic damage as it is seen in early diabetes.
"Summary
Background: The presence of autonomic dysfunction in HIV patients is largely unknown. Early studies found autonomic dysfunction in patients with AIDS. Introduction of highly active antiretroviral combination therapy (ART) has dramatically changed the course of the disease and improved prognosis and decreased morbidity. At present it is not known whether introduction of ART also has decreased autonomic dysfunction.
Aim: To evaluate whether autonomic dysfunction is present in an ART-treated HIV population.
Methods: HIV patients receiving ART for at least 3 years (n = 16) and an age-matched control group of healthy volunteers (n = 12) were included. All were non-smokers, non-diabetic and had never received medication for dyslipidaemia or hypertension. Following a 10 min resting period a 5 min ECG recording was performed. Heart rate variability (HRV) analysis was performed in accordance with current guidelines and data reported as median (interquartile range).
Results:
The resting heart rate was higher in HIV patients compared with controls [69 (62-74) versus 57 (52-60); P<0·001].
Total HRV measured as standard deviation of normal-to-normal (SONN) was lower in the HIV group compared with the controls [36 (25-55) versus 74 (57-84) ms; P<0·01] as was parasympathetic activity measured as square root of the mean squared difference of successive normal-to-normal intervals (RMSSD) [22 (9-30) versus 35 (24-62) ms; P<0·05].
Low frequency power was lower in the HIV group compared with the control group [294 (161-602) versus 946 (711-1668) ms2; P<0·01]. High frequency power as well as systolic and diastolic blood pressure did not differ between the groups.
Conclusions: The HIV patients in ART have increased resting heart rate and decreased short-term heart rate variability indicating parasympathetic dysfunction.
Introduction
Autonomic dysfunction has been shown to be common in patients with HIV and AIDS. The prevalence seems to be particularly high in patients with advanced disease (Ruttimann et al., 1991; Welby et al., 1991). Most studies have been performed in untreated patients or patients with very advanced disease. However, the availability of potent combination antiretroviral regimes has resulted in a dramatic reduction in HIV-associated morbidity and mortality in the developed world. It would therefore be of interest to study whether such highly active antiretroviral therapy (HAART) has influenced the prevalence of autonomic dysfunction in an optimally treated HIV population. Suppression of HIV during ART could reduce the prevalence of autonomic dysfunction. However, ART is also known to induce an array of adverse effects, among them dyslipidaemia and insulin resistance (Carr et al., 1998; Dube & Sattler, 1998; SoRelle, 1998). As diabetes is a known cause of autonomic dysfunction the diabetogenic effect of ART could cause autonomic dysfunction in HIV patients in ART.
Most previous studies of autonomic function in HIV patients have used simple autonomic function tests. However, the use of heart rate variability (HRV) with spectral analysis allows for the assessment of relative contribution of sympathetic and parasympathetic involvement.
The aim of the present study was therefore to study the presence of autonomic dysfunction in a population of optimally treated HIV positives receiving ART. To do so, we performed analysis of HRV to assess the relative contribution of the sympathetic and parasympathetic nervous system.
Material and methods
Subjects
Between January 2004 and November 2004 patients and controls were enrolled. The Danish Scientific Ethical Committee approved the study and written informed consent was obtained from all the participants.
HIV patients
A total of 16 patients with HIV infection were included. The patients were recruited from the outpatient clinic at the Department of Infectious Diseases, Rigshospitalet or Department of Infectious Diseases, Hvidovre Hospital in Copenhagen, Denmark. All were Caucasians.
Inclusion criteria were: (i) age 18-60 years, (ii) proven HIV infection, (iii) receiving stable doses of antiretroviral medication for ≥36 months and (iv) non-smoking. Exclusion criteria: (i) clinical history of cardiovascular or ischaemic heart disease, (ii) known alcohol abuse, (iii) lipid-lowering medication, (iv) known diabetes mellitus, (v) medication for hypertension and (vi) pregnancy. None of the patients had a family history of coronary heart disease or stroke. The patients had been HIV positive for 4-20 years (median: 12 years). All patients had received ART for more than 44 months. Four (25%) of the HIV patients had AIDS. Patient characteristics are shown in Table 1.
Controls
Twelve non-smoking age- and sex-matched healthy volunteers were included. Their basic characteristics are summarized in Table 1. The control subjects were included from individuals participating in a Danish prospective population study (Copenhagen Center of Prospective Population Studies at the Institute of Preventive Medicine, Bispebjerg Hospital, University of Copenhagen, Denmark). All were Caucasians.
Heart rate variability testing
Following a 10-min resting period a 5 min ECG recording was performed while patients were in the supine position. HRV analysis was performed in accordance with current international guidelines (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The ECG data were sampled at 1·0 kHz and analysed using a commercially available program (Chart v. 4.2.3 with HRV module v. 1; AD Instruments, Colorado Springs, CO, USA) that complies with current guidelines. The time domain analysis variables were: mean normal-to-normal (NN; units: ms), standard deviation of NN (SDNN; units: ms) and square root of the mean squared difference of successive NN-intervals (RMSSD; units: ms). SDNN is considered an estimate of overall HRV and thus both an indicator of sympathetic and parasympathetic influence. RMSSD is an estimate of short-term components of HRV and thus considered reflecting the parasympathetic influence. The frequency domain analysis (spectral analysis) was performed by fast Fourier transformation using a window of 1024 points, with a standard Welch filter and the variables were: low frequency (LF) power (0·04-0·15 Hz; units: ms2), LF power in normalized units (LFnu) which is calculated as LF / (total power-very LF power) x 100, high frequency (HF) power (0·15-0·4 Hz; units: ms2), HF power in normalized units (HFnu) which is calculated as HF / (total power-very LF power) x 100 and the ratio between LF and HF power (LF / HF ratio).
Statistical analysis
As most of the variables were not normally distributed data are presented as median with 25 and 75 percentiles in parenthesis. Comparisons between the groups were performed using either Fisher's exact test for binomial data (2 x 2 contingency tables) or Mann-Whitney test for continuous data. P<0·05 was considered significant.
Results
The HIV group and control group did not differ with respect to age, gender, body mass index or systolic and diastolic blood pressure (Table 1).
Heart rate variability: time domain analysis
The resting heart rate was significantly higher in HIV patients compared with controls (69 versus 57 bpm; P<0·001). Accordingly mean NN was significantly lower in HIV patients compared with the controls (855 versus 1103 ms; P<0·001). SDNN was significantly lower in the HIV group compared with the controls (36 versus 74 ms; P<0·01) as was RMSSD (22 versus 35 ms; P<0·05). Data of the time domain analysis are summarized in Table 2.
Heart rate variability: frequency domain analysis (spectral analysis)
Low frequency power was significantly lower in the HIV group compared with the control group (294 versus 946 ms2; P<0·01) whereas no difference was found in LFnu between the groups. HF power, HFnu or LF / HF ratio did not differ between the groups. Data of the frequency domain analysis are summarized in Table 3.
Discussion
Our finding of increased resting heart rate and corresponding decreased mean NN in HIV patients indicate a change in parasympathetic / sympathetic balance. It could in theory be due to either lower parasympathetic and / or higher sympathetic tone in HIV patients compared with controls although lower parasympathetic influence is most likely due to its stronger influence on resting heart rate. The time domain analysis of HRV revealed a decrease in both SDNN and RMSSD. SDNN reflects overall HRV and thus a decrease in HIV patients indicates an overall decrease in combined parasympathetic and sympathetic modulation of heart rate. RMSSD is an indicator of short-term components of HRV and thus is considered to reflect the degree of parasympathetic modulation of the heart rate. Therefore, the lower values of RMSSD found in the HIV patients indicate an impaired parasympathetic influence on heart rate. The frequency domain analysis revealed a decrease in LF indicating a possible decrease in the combined parasympathetic and sympathetic modulation. When normalized, there was no difference in LFnu indicating that the lower LF power in HIV patients was due to a lower total power rather than a lower sympathetic contribution. HF power was not significantly lower in the HIV patients, which might be surprising as HF power is an indicator of parasympathetic tone and therefore is in contrast to the finding of a lower RMSSD in the HIV patients. However, the variation in HF power was much larger than in RMSSD and the latter may be the most reliable measurement for parasympathetic activity. Indeed, there was a tendency to lower HF power in the HIV group (P = 0·2) but the risk of statistical type II error was substantial due to the relative limited number of patients. Taken together, the overall picture clearly indicates an increase in resting heart rate and a decrease in HRV mainly because of a decreased parasympathetic influence.
As the age and gender match was not perfect between our study groups, it could be argued that this may also have contributed to the differences found. However, we performed a regression analysis with age, gender and HIV versus not HIV as independent variables. In all cases only HIV was a significant parameter indicating that the observed differences in autonomic parameters between the groups are indeed explained by HIV.
Most previous studies of autonomic function in HIV patients have been performed using simple tests of autonomic function (Snider et al., 1983; Craddock et al., 1987; Freeman et al., 1990; Scott et al., 1990; Ruttimann et al., 1991; Welby et al., 1991; Rogstad et al., 1999; Nzuobontane et al., 2002) and only few studies have used HRV testing. The advantage of HRV testing is the ability to discriminate, at least in part, between the parasympathetic and sympathetic influence (Eckberg, 1997). In one of the studies using HRV analysis in 10 AIDS patients with advanced disease it was found that HIV patients had profoundly decreased values of all frequency domain components indicating decreased sympathetic as well as parasympathetic influence (Neild et al., 2000). However, compared with our patients, the study was performed on patients with much more advanced disease than in our study and none received HAART. In another study of 21 treatment naïve HIV patients, reductions were found in the time domain analysis of mean NN, SDNN and RMSSD (Mittal et al., 2004). In the frequency domain analysis, reductions were found in total power, LF power and HF power but not LFnu or HFnu. It was concluded that both the sympathetic and parasympathetic component were affected. In a study that included a group of non-treated HIV patients, no difference in resting heart rate was observed but frequency domain analysis revealed a decrease in LFnu but not in HF power (Correia et al., 2006). The time domain variables SDNN and RMSSD were not reported. Using a cold-face test and tilt test it was concluded that the patients responded with a decreased parasympathetic and an increased sympathetic response, respectively. It should be noted that in all the above-mentioned studies subjects were substantially younger than in our study.
However, today the most relevant group of HIV patients in the Western World is patients receiving ART. Since the introduction of ART 10 years ago it has been possible to treat HIV patients and when treated optimally to obtain continuous suppression of HIV RNA as well as high CD4-cell counts, which indicates immune competence. The treatment has decreased mortality of HIV patients but has also lead to long-term concern about possible adverse effects of treatment. Adverse effect could be both due to the drugs themselves, as well as caused indirectly through development of hypercholesterolaemia, insulin resistance and metabolic syndrome, well known to be associated with ART (Carr et al., 1998; Dube & Sattler, 1998; SoRelle, 1998). Development of parasympathetic autonomic dysfunction may be especially dangerous if patients at the same time develop ischaemic heart disease as the combination may be arrhythmogenic. Very limited data are available on autonomic function and HRV in HIV patients on ART. To our knowledge, the only study that exists at present studied a group of 40 HIV positive patients being ART naïve, 40 AIDS patients on ART and 40 controls (Correia et al., 2006). No autonomic dysfunction was found in the group of AIDS patients on ART when compared with the control group. Compared with our study group, the patients were much younger, had been shorter on ART (6 months+), and were less optimally treated as only 9 / 40 had HIV load <50 copies ml-1 and a significantly lower mean CD4 cell count. We believe that our population of HIV patients, with a disease duration of 4-20 years and treated with ART for at least 3 years (median 7 years) is highly relevant for judgment of possible autonomic dysfunction in an optimally treated Western World HIV population. The long-standing treatment makes it likely that any treatment-induced long-term adverse effects on the autonomic nervous system should be demonstrable. Indeed, we found a clear and selective parasympathetic dysfunction with no indication of sympathetic dysfunction. Compared with studies of advanced disease in untreated patients where in general both sympathetic and parasympathetic dysfunction were present, it may be speculated whether the dysfunction observed in our population has a different pathogenesis than the one observed in the pretreatment era. The autonomic dysfunction in untreated patients with advanced disease is generally believed to be caused by HIV-1 virus itself, which is well known to be neurotropic (Ruttimann et al., 1991; Melli et al., 2006) In contrast, the selective parasympathetic dysfunction found in our patients could rather be caused by the anti-retroviral drugs used that are known to be neurotoxic. Accordingly, some of the nucleoside reverse transcriptase inhibitors especially stavudine are known to affect mitochondrial function and may lead to depletion of the neuronal mitochondrial DNA and to an inhibition of DNA polymerase (Dalakas et al., 2001; Peltier & Russell, 2002; Authier & Gheradi, 2003). These effects on the mitochondria may be related to the polyneuropathy that develops in some of the patients receiving these drugs. Although a total of eight of our patients had received stavudine previously, only one patient was on the treatment when the examination was performed. As all the patients did have a lactate in the normal range, we do not believe that stavudine induced hyperlactataemia is the reason for our findings. The diabetogenic effect of ART could lead to parasympathetic damage as it is seen in early diabetes.
A regression model with HIV duration, AIDS, CD4, CD4 nadir and HAART duration as independent variables did not find any correlation with these and the autonomic parameters. It should however be noted that our study was not designed to study this, e.g. did we not include patients with short disease or HAART duration.
Conclusion
HIV patients in long-standing HAART have increased resting heart rate and decreased short-term HRV indicating parasympathetic dysfunction.
Study limitations
Since we performed a cross-sectional study, we cannot exactly define whether the autonomic changes seen were induced de novo during ART or were already present prior to therapy. In theory, these patients could have had combined sympathetic and parasympathetic dysfunction prior to ART and had recovered from sympathetic dysfunction during treatment although we find this most unlikely as sympathetic neuropathy is generally considered irreversible. The exact time course of development of autonomic dysfunction in relation to ART should be established in future longitudinal studies.
| |
| |
| |
|
|
|
