We describe a “gold standard” for evaluating orthostatic tolerance (OT) using tilt testing with combined lower body negative pressure (LBNP). This can be combined with non-invasive evaluations of cardiovascular reflex control. Normal and abnormal responses are defined.
Orthostatic tolerance (OT) refers to the ability to maintain cardiovascular stability when upright, against the hydrostatic effects of gravity, and hence to maintain cerebral perfusion and prevent syncope (fainting). Various techniques are available to assess OT and the effects of gravitational stress upon the circulation, typically by reproducing a presyncopal event (near-fainting episode) in a controlled laboratory environment. The time and/or degree of stress required to provoke this response provides the measure of OT. Any technique used to determine OT should: enable distinction between patients with orthostatic intolerance (of various causes) and asymptomatic control subjects; be highly reproducible, enabling evaluation of therapeutic interventions; avoid invasive procedures, which are known to impair OT1.
In the late 1980s head-upright tilt testing was first utilized for diagnosing syncope2. Since then it has been used to assess OT in patients with syncope of unknown cause, as well as in healthy subjects to study postural cardiovascular reflexes2-6. Tilting protocols comprise three categories: passive tilt; passive tilt accompanied by pharmacological provocation; and passive tilt with combined lower body negative pressure (LBNP). However, the effects of tilt testing (and other orthostatic stress testing modalities) are often poorly reproducible, with low sensitivity and specificity to diagnose orthostatic intolerance7.
Typically, a passive tilt includes 20-60 min of orthostatic stress continued until the onset of presyncope in patients2-6. However, the main drawback of this procedure is its inability to invoke presyncope in all individuals undergoing the test, and corresponding low sensitivity8,9. Thus, different methods were explored to increase the orthostatic stress and improve sensitivity.
Pharmacological provocation has been used to increase the orthostatic challenge, for example using isoprenaline4,7,10,11 or sublingual nitrate12,13. However, the main drawback of these approaches are increases in sensitivity at the cost of unacceptable decreases in specificity10,14, with a high positive response rate immediately after administration15. Furthermore, invasive procedures associated with some pharmacological provocations greatly increase the false positive rate1.
Another approach is to combine passive tilt testing with LBNP, providing a stronger orthostatic stress without invasive procedures or drug side-effects, using the technique pioneered by Professor Roger Hainsworth in the 1990s16-18. This approach provokes presyncope in almost all subjects (allowing for symptom recognition in patients with syncope), while discriminating between patients with syncope and healthy controls, with a specificity of 92%, sensitivity of 85%, and repeatability of 1.1±0.6 min16,17. This allows not only diagnosis and pathophysiological assessment19-22, but also the evaluation of treatments for orthostatic intolerance due to its high repeatability23-30. For these reasons, we argue this should be the “gold standard” for orthostatic stress testing, and accordingly this will be the method described in this paper.
Throughout testing, continuous beat-to-beat blood pressure and electrocardiogram (ECG) monitoring is paramount. This ensures subject safety, and prompt termination of the test with the onset of presyncope. Beat-to-beat blood pressure recordings can be obtained through arterial catheterization, or finger plethysmography31-33. The latter is used in this protocol because it is non-invasive and can assess the onset of presyncope with the same accuracy as catherization31,34, without the detrimental impact of invasive monitoring on OT1. Using the Modelflow technique changes in stroke volume, cardiac output, and total peripheral resistance can be derived from the finger arterial pressure waveform35,36. Additional noninvasive measures that may aid the haemodynamic evaluation can also be conducted, and will be described here. Continuous end tidal oxygen (PETO2) and carbon dioxide (PETCO2) monitoring using a nasal cannula allows the evaluation of any contribution of hyperventilation to the subject’s symptoms. Finally, monitoring both brachial and cerebral blood flow velocities using Doppler ultrasound can be undertaken to allow the determination of peripheral and cerebral vascular responses to orthostasis. In addition, measurements of venous pooling and capillary filtration could also be obtained using impedance plethysmography20. Ultimately, this protocol allows assessment of postural cardiovascular reflex control in a controlled and reproducible setting.
1. Equipment
2. Data Collection
Using this protocol, all subjects experience presyncope, and the definition of normal or abnormal responses is made largely based upon the time it takes to induce this reaction. OT is defined as the time to presyncope in minutes from the onset of upright tilting. Typical values for OT in healthy volunteers according to age and gender can be seen in Table 1. Patients with orthostatic intolerance exhibit presyncope earlier in the test, with 85% ending the test within the -20 mmHg phase compared to 23% of controls17. The threshold of normal is taken as the 20% incidence of syncope17.
At termination of the test the haemodynamic responses fall into essentially three categories: vasovagal syncope; postural tachycardia syndrome (POTS); or autonomic failure46. The vasovagal response is characterized as sudden onset hypotension and bradycardia, although the relative contributions of each component can vary greatly, and have been characterized according to the VASIS classification46,47. A representative response in an adult control is depicted in Figure 5. At tilt, the baroreflex compensates for blood pooling in the lower extremities through vasoconstriction, reflected in the reduced brachial blood flow and increased diastolic pressure, and tachycardia. This response is maintained until presyncope, at which point there is a sudden switch to vasodilatation and bradycardia. In this case the subject experienced a vasovagal response at presyncope, associated with symptoms of dizziness and warmth. This is the most common response to tilting in healthy controls.
In patients with autonomic failure there is an absence of adaptation of blood pressure to the upright position, with a slow and progressive decline in blood pressure, and small or absent heart rate response to orthostatic stress. Patients will develop symptoms once a critical level of blood pressure is reached (typically at or below a systolic pressure of 80 mmHg46, although many patients with autonomic failure tolerate much lower blood pressures than this) representing the lower limit for cerebral autoregulation. This response generally occurs only in older patients with syncope, who often present with other associated disorders46.
POTS is characterized by a heart rate rise of 30 bpm within 10 min of head-upright tilting, or with an upright heart in excess of 120 bpm48,49, associated with symptoms of presyncope. Blood pressure is usually well maintained. This disorder generally affects young females (<40 years old), and typically presents with presyncope, and only occasionally with syncope48. Heart rate responses to orthostatic stress are brisk in children and adolescents, so alternative diagnostic criteria for POTS are recommended in this population50.
Regardless of the haemodynamic response, there is a drop in cerebral blood flow velocity at presyncope51. There may be impairments in cerebral autoregulation in patients with vasovagal syncope22 and POTS52.
In those with intact baroreflex control of vascular resistance there will be a progressive reduction in brachial blood flow velocity during the orthostatic stress, associated with increased forearm vascular resistance. In healthy controls a maximum vascular resistance response of +100±12% is considered normal21. Smaller responses are indicative of impaired vascular resistance responses19,21, with 60% of patients with presyncope and poor OT having maximal vascular resistance responses below +80%, and 83% of controls having responses that exceed this value19. Patients with POTS often have particularly small peripheral vascular resistance responses21. There is a significant positive correlation between the maximum vascular resistance response and OT19.Typical heart rate responses in healthy controls at the end of the head up tilt phase and peak heart rates during the orthostatic stress are 71-82 bpm and 98-133 bpm respectively16,21.
Some degree of hyperventilation is common during orthostatic stress, with associated reductions in PETCO253. This tends to produce a hypocapnic cerebral vasoconstriction and peripheral vasodilation, and so contributes to the decreases in blood pressure and cerebral blood flow at presyncope. Some individuals are excessively sensitive to alterations in PETCO2 and this may be a factor in predisposing them to syncopal events53.
Figure 1. The manual, adjustable head-up tilt table and LBNP chamber. The manual design allows for rapid transitions between supine and tilt, ensuring the prompt recovery of presyncopal symptoms at test termination. A. The handle used to move the table into different positions. B. A pressure gauge monitors the level of LBNP during the protocol. C. A wooden waist board with neoprene is attached to the top of the chamber to ensure the chamber is air-tight. D. The adjustable arm rest with attached clamp for the brachial ultrasound probe. E. A seatbelt over the subject’s legs minimizes skeletal muscle pumping activity during orthostatic stress. F. The lever used to adjust the table from -15 to 60 °. G. The footplate can be adjusted for the subject’s height once their iliac crest is positioned in the center of the table.
Figure 2. Head-up tilt protocol. There are three different phases in the protocol: supine for 20 min; tilt for 20 min; and LBNP for 10 min each at -20 mmHg, -40 mmHg, and -60 mmHg.
Figure 3. Brachial blood flow velocity measurement. A. The brachial ultrasound probe is positioned overlying the brachial artery to enable measurements of forearm blood flow velocity, and the calculation of vascular resistance responses. B. Once in place, the probe is secured using an adjustable clamp to ensure the angle of insonation does not change throughout the test.
Figure 4. Middle cerebral artery blood flow velocity measurement. A. A 2MHz ultrasound probe is positioned on the transtemporal window overlying the middle cerebral artery. An example showing middle cerebral artery blood flow velocity is shown in the top panel. The birfurcation point, with the anterior cerebral artery waveform just visible as a negative deflection, enables confident identification of the middle cerebral artery. The bottom panel shows an example from the posterior cerebral artery. Although the flow profile is similar to the middle cerebral artery, the mean velocity is lower and insonation depth greater, enabling ease of discrimination between the two vessels. B. Once in place the probe is secured in position using either a plastic headset (upper) or fabric headbands (lower).
Figure 5. Example tracing from an adult control volunteer. Beat-to-beat blood pressure (BP), electrocardiogram (ECG), brachial (BBFV) and cerebral (CBFV) blood flow velocity are shown. During the orthostatic stress a progressive baroreflex mediated tachycardia and vasoconstriction can be seen. At presyncope, the baroreflex begins to fail and there is a drop in blood pressure and cerebral blood flow velocity.
Orthostatic Tolerance (min) | ||||||
Males | Females | |||||
Age (years) | 20-35 | 36-50 | >50 | 20-35 | 36-50 | >50 |
Mean | 35±1.4* | 35±1.7 | 35±1.7 | 29±1.5 | 31±1.8 | 37±2.3 |
20% incidence | 30 | 34 | 32 | 24 | 28 | 30 |
50% incidence | 34 | 37 | 36 | 29 | 32 | 38 |
Table 1. Predicted times to presyncope in healthy volunteers according to age and gender. Values are given for the mean orthostatic tolerance ± standard error, as well as the times at which 20% and 50% of individuals experienced presyncope. * denotes significant difference (p<0.05) between males and females within the age group. There was a significant correlation between age and onset of presyncope in females (r=0.63; p=0.004), but not males. Combined data (n=63) from El-Bedawi and Hainsworth (1994), and Claydon, unpublished observations.
This technique is highly reproducible, has the ability to discriminate normal and abnormal responses with high sensitivity and specificity, and can provoke presyncope in all subjects, allowing for symptom recognition in patients with recurrent syncope. In a clinical setting, different types of syncope can be distinguished, allowing tailored treatment and management approaches. The impact of interventions can readily be assessed. With additional cardiovascular monitoring, reflex responses can also be evaluated.
In order to differentiate certain sub-types of syncope in patient populations, the analysis of plasma catecholamine responses during orthostatic stress may be useful. However, the invasive nature of this procedure may interfere with the accurate determination of OT1.
This protocol can easily be modified to account for specific clinical or research questions, or the needs of specific populations. The angle of the table can be altered in situations when 60 ° may be difficult to sustain, for example when evaluating orthostatic intolerance in spinal cord injured populations. In such cases, tilting to 30-35 ° may be more appropriate and tolerable54-56. If desired, the protocol may be shortened in specific populations by stopping after completion of LBNP at -20 mmHg – individuals who reach the end of this phase essentially have a normal OT (Table 1). This may be applicable if the goal is to determine whether the response is abnormal or not, but not necessarily to define the precise OT. This may be particularly pertinent in paediatric populations.
Often with an automatic table, the transition back to supine is slow and this can increase the likelihood of profound bradycardia or asystole, and brief loss of consciousness, occurring coincident with a vasovagal response57. As such, the benefit of the manual table is that the pathophysiological and symptomatic responses associated with presyncope are quickly recovered.
An experienced physiologist, technician or nurse must be present during testing. There is a lack of agreement as to whether a medical doctor should also be present or available (on call) in case of asystole or other complications3,6,58. The laboratory should have access to atropine should asystole occur, as well as a defibrillator. Serious adverse events are rare and no deaths have been reported58. Brief periods of profound bradycardia or asystole may accompany vasovagal responses4,24,46, but they typically recover spontaneously with the return to a supine posture. Seizure-like activity may accompany syncope59, but can be avoided with termination of the test at presyncope. There is one report of cardiopulmonary resuscitation being required in the case of tilt-induced coronary vasospasm in a patient with known coronary vasospastic angina60.
The authors have nothing to disclose.
We would like to acknowledge Professor Roger Hainsworth, who developed this technique. We are grateful to Mr. King Hang Chao and Mr. Wang-Joe Woo for their assistance with photography.
This work is supported by Simon Fraser University and the Heart and Stroke Foundation of Canada.
Equipment | Manufacturer | Location |
Tilt Table | Custom-build | Leeds, United Kingdom |
Finometer | Finapres Medical Systems | Amsterdam, The Netherlands |
Doppler Box | Compumedics | Singen, Germany |
Doppler software | The DWL Doppler Company | Singen, Germany |
Aquasonic Ultrasound gel | Parker Laboratories, Inc. | Fairfield, USA |
Headbands | Lululemon | Burnaby, Canada |
Headset | Canadian Tire | Burnaby, Canada |
ECG | Finapres ECG Module, Finapres Medical Systems | Amsterdam, The Netherlands |
Electrodes | Red Dot | Ontario, Canada |
Antiseptic Isopropyl Alcohol Pads | Lernapharm | Quebec, Canada |
O2Cap-Oxygen Analyser | Oxigraph Inc. | California, USA |
Airlife Nasal Oxygen Cannula | Cardinal Health | Mountainview, USA |
Powerlab 16/30 | AD Instruments | Colorado Springs, USA |