Blood pressure oscillations during tilt testing as a ...

5 downloads 35464 Views 237KB Size Report
GTN provocation ('responders') were assigned oscillation-positive or oscillation-negative ... Tel: þ44 207 380 9776, Fax: þ44 207 388 5095, Email: [email protected] ..... manifestation of a failing auto-regulatory mechanism that may con-.
CLINICAL RESEARCH

Europace (2009) 11, 1696–1701 doi:10.1093/europace/eup338

Syncope and implantable loop

Blood pressure oscillations during tilt testing as a predictive marker of vasovagal syncope Derek John Hausenloy 1*, Chanpreet Arhi 2, Navin Chandra3, Ann-Christine Franzen-McManus 4, Andrea Meyer 4, and Richard Sutton 4 1 The Hatter Cardiovascular Institute, University College London Hospital and Medical School, 67 Chenies Mews, London WC1E 6HX, UK; 2Chelsea and Westminster Hospital, 369 Fulham Road, London SW10 9NH, UK; 3King’s College Hospital, Denmark Hill, London SE5 9RS, UK; and 4Imperial College Healthcare NHS Trust, St Mary’s Hospital, Praed Street, London W2 1NY, UK

Received 18 June 2009; accepted after revision 3 October 2009; online publish-ahead-of-print 31 October 2009

Aims

----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords

Head-up tilt † Vasovagal syncope † Blood pressure † Oscillations

Introduction Accurate diagnosis of vasovagal syncope relies on a thorough medical history and physical examination and head-up tilt (HUT) testing.1 Drug provocation using either isoprenaline (Iso) or, more recently, glyceryl trinitrate (GTN) has been used to increase the sensitivities of HUT to 61 and 69%, respectively, with specificities of 92 and 94%, respectively.2,3 Overall, sublingual GTN, as part of the Italian protocol (a 0.4 mg GTN spay after a 20 min passive tilt phase),2 is recommended as the preferred initial agent to provoke syncope due to its better tolerability, its non-invasive administration, and its comparable sensitivity and specificity when compared with Iso.2 Some studies have investigated the combined

administration of GTN and Iso and have reported further increases in sensitivity to 79–88% at the expense of a reduction in specificity to 67%.4,5 In order to elucidate the pathophysiological mechanisms of vasovagal syncope, studies have examined the haemodynamic events which precede the onset of syncope in susceptible individuals and have classified patients, according to the Vasovagal Syncope International Study,6 into distinct subgroups based on the characteristic changes in heart rate and blood pressure which herald the onset of syncope.7 Interestingly, recent studies have observed a phase of haemodynamic instability immediately preceding the sudden fall in blood pressure and heart rate, which signal the onset of syncope and during which there are

* Corresponding author. Tel: þ44 207 380 9776, Fax: þ44 207 388 5095, Email: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2009. For permissions please email: [email protected].

Downloaded from by guest on November 15, 2015

During head-up tilt (HUT) testing, a period of haemodynamic instability, marked by oscillations in blood pressure, often precedes vasovagal syncope. We hypothesized that the presence of oscillations in blood pressure during HUT testing predicts a positive diagnosis for vasovagal syncope. ..................................................................................................................................................................................... Methods The haemodynamic profiles of 42 consecutive patients non-responsive to passive HUT and glyceryl trinitrate (GTN) and results provocation (‘non-responders’) and, contemporaneously, 41 consecutive patients responsive to passive HUT and GTN provocation (‘responders’) were assigned oscillation-positive or oscillation-negative depending on the presence or absence of a characteristic oscillation in systolic blood pressure which varied by 30 mmHg (peak-to-trough). All the non-responders proceeded to an isoprenaline (Iso) challenge test. Of the 42 non-responders, 27 patients were Iso tilt-positive; all of these patients were assigned oscillation-positive. The other 15 non-responders were Iso tiltnegative; of these 9 were assigned oscillation-positive and 6 were assigned oscillation-negative. Of the 41 responder patients, 33 were assigned oscillation-positive, whereas 8 were assigned oscillation-negative. Overall, the presence of oscillations as a diagnostic predictor for vasovagal syncope had a sensitivity of 88% (positive predictive value of 87%) and a specificity of 40% (negative predictive value of 43%). ..................................................................................................................................................................................... Conclusion In patients non-responsive to passive HUT and GTN provocation, the presence of an oscillating systolic blood pressure varying 30 mmHg may still indicate a diagnosis of vasovagal syncope, possibly obviating the need for Iso testing.

1697

Blood pressure oscillations during tilt testing

oscillations in both blood pressure and heart rate at a frequency of 0.1 Hz.8 The actual cause of this haemodynamic instability is unclear although it has been attributed to attenuated baroreceptor sensitivity,8,9 and possible variations in sympathetic activity occurring on a background of low parasympathetic activity that it is unable to buffer the oscillations.8 In engineering terms, this is a servo-controlled mechanism under stress, constantly self-correcting. In our hospital, patients with suspected vasovagal syncope that are non-responsive to passive HUT and GTN provocation, in whom a diagnosis of vasovagal syncope is still suspected, are subjected to Iso provocation on a separate occasion. Given the side effects associated with Iso provocation described earlier, the aim of this study was to determine whether it is possible to predict a subsequent positive response to Iso provocation from the analysis of the haemodynamic response to HUT in patients who were not syncopal (non-responder) to passive HUT and GTN provocation, thereby possibly obviating the need for Iso testing in this patient group.

Methods This retrospective study involved analysing the haemodynamic response of two groups of patients: (i) non-responders: 42 consecutive subjects with recorded data of adequate quality for analysis, referred with suspected vasovagal syncope who had previously undergone passive HUT and GTN provocation with no reproduction of symptoms, and who had gone on to receive Iso provocation at a later date and (ii) responders: 41 consecutive contemporaneous patients with recorded data of adequate quality for analysis, referred with suspected vasovagal syncope in whom passive HUT with or without GTN provocation did reproduce symptoms. Patients investigated for vasovagal syncope were recruited in a consecutive manner between November 2003 and November 2005.

The two-stage head-up tilt protocol with sublingual glyceryl trinitrate The HUT test was performed by means of an electrically controlled tilt table with a footboard for weight-bearing. Blood pressure, heart rate, and heart rhythm were continuously monitored and recorded. Heart rate data were obtained as a series of R – R intervals with resolution of 1 ms. Blood pressure was measured by means of a Finapres 2300 (Ohmeda, Louisville, CO, USA) blood pressure photoplethysmographic device, which provides continuous non-invasive beat-to-beat determination of finger arterial pressure by the volume-clamp method, and which accurately reflects changes.10 The HUT test was performed after an initial observation with the patient in the supine position for 10 min. The test comprised two consecutive stages: in stage 1, patients were tilted at 608 for up to 20 min without medication. If syncope (or limiting symptoms) did not develop, patients entered stage 2, in which they received 0.4 mg of sublingual GTN and continued to be tilted for a further 15 min. If syncope (or limiting symptoms) occurred during the test, the tilt table was rapidly lowered to return the patient to the supine position, and the study was terminated. Patients non-responsive to passive HUT and GTN provocation, in whom a diagnosis of vasovagal syncope was still clinically suspected, received Iso provocation on a separate day.

Patients undergoing passive HUT and GTN provocation in which syncope had not been reproduced, underwent Iso provocation on a separate occasion. An intravenous line with a small gauge cannula was inserted at least 20 min before the test. A resting period of 10 min was recorded in the supine position, before the patient was tilted at 608 for 10 min. Following this, an Iso infusion was started at 1 mg/min for 5 min. If there was no syncope, the infusion rate was increased to 2 mg/min for a further 5 min. The infusion was increased by 1 mg/min every 5 min while keeping the patient tilted, repeated to a maximum of 5 mg/min, the aim being to precipitate vasovagal syncope and reproduce the patient’s usual symptoms. If, at this point, there was no syncope, the infusion was discontinued and the patient was returned to the supine position. In addition, the infusion was stopped if there was an increase in ventricular ectopics or runs of ventricular ectopics or there were limiting side effects. The patient was then monitored until heart rate and blood pressure had returned to normal at which stage the recording was stopped. Patients were warned beforehand that Iso may cause palpitations, dry mouth, headache, a feeling of apprehension, warmth, and sweatiness. They were advised that they should try to distinguish between symptoms experienced as a result of the Iso infusion from those which are the consequence of syncope. This was aided by careful questioning of the patients concerning the relevance of symptoms generated to those experienced in every day life.

Data collection and processing Data were obtained for a supine rest period of 10 min, throughout the period of the tilt, during the course of the drug provocation and for 2 min after return to the supine position. Data from each patient were reviewed and edited manually to remove artefacts. Then, heart rate and blood pressure were averaged for the rest period, for each minute of the tilt, and for the post-tilt period.

Analysis of oscillations For the purposes of the study, a 30 mmHg peak-to-trough altered pressure variation between two consecutive cycles was considered to be oscillation. A figure of 30 mmHg was chosen as it has been demonstrated in a previous study that 0.1 Hz oscillations in peak-to-trough systolic blood pressure with an average value of 27 mmHg occur in patients who go on to have a vasovagal syncope.8 In almost all cases, this was sustained over many more cycles.

Data statistical analysis Values are presented as mean + SD. Student’s unpaired t-test was used to exclude a significant difference in the age of the different patient groups.

Results There were 42 patients (21 male and 21 female) who undertook the passive HUT and GTN provocation with no reproduction of syncope, and went on to Iso provocation at a later date (assigned non-responders). Forty-one patients (10 male and 31 female) undertook the passive HUT and GTN provocation with reproduction of syncope (assigned responders).

Downloaded from by guest on November 15, 2015

Subjects

Isoprenaline provocation

1698

Isoprenaline testing Of the 42 non-responders referred for Iso testing, 27 (71%) had a positive test with reproduction of pre-syncopal or syncopal symptoms and 15 (29%) had a negative test with no reproduction of pre-syncopal or syncopal symptoms. The mean age for the Isopositive group was 39 + 18 years (48% male, 52% female) and for the Iso-negative group was 56 + 19 years (62% male, 38% female), this difference was not significantly different.

Analysis of the blood pressure response The haemodynamic response to passive HUT and GTN provocation was analysed in both the non-responders and responders. The difference between the maximum and minimum systolic blood pressure achieved during the test was noted for the two stages of the HUT protocol, i.e. passive HUT and following GTN provocation. Since oscillation in the supine phase, pre-tilt, was also seen in some patients, we included the baseline period in the analysis.

30 mmHg, while the remaining six (14%) did not display the same degree of haemodynamic instability as indicated by a relatively steady blood pressure recording throughout the testing (assigned oscillation-negative) (see Figure 2 for representative example). The mean age of the oscillation-positive group was 43 + 20 years compared with 56 + 20 years for the oscillation-negative group (P . 0.05). Of the 36 patients who were noted to be oscillation-positive, 27 (75%) went on to have a positive Iso test (mean dose 3.9 + 0.9 mg/ min). The remaining nine (25%) oscillation-positive patients went on to have a negative Iso test. All six of the oscillation-negative patients went on to have a negative Iso test, with no patients having a positive Iso test.

Responders to passive head-up tilt and glyceryl trinitrate After analysis by an operator, 33 of the 41 responder patients (81%) were observed to display significant haemodynamic instability, indicated by oscillations of blood pressure (assigned oscillationpositive), and a variation of systolic blood pressure 30 mmHg, while the remaining eight patients (19%) did not display the same degree of haemodynamic instability as indicated by a relatively steady blood pressure recording throughout the testing (assigned oscillation-negative). The mean age of the oscillationpositive group was 50 + 21 years compared with 66 + 17 years for the oscillation-negative group (P . 0.05).

Figure 1 Representative chart recording of a typical oscillation-positive patient demonstrating the characteristic oscillating systemic blood pressure trace with variations in systolic blood pressure of 30 mmHg (peak-to-trough) occurring during either the passive tilt phase or during the glyceryl trinitrate (GTN) provocation.

Downloaded from by guest on November 15, 2015

Non-responders to passive head-up tilt and glyceryl trinitrate After analysis by an operator blinded to the subsequent Iso test result, 36 of the 42 patients (86%) were noted to display significant haemodynamic instability, indicated by oscillations of blood pressure (assigned oscillation-positive) (see Figure 1 for representative example), and a variation of systolic blood pressure

D.J. Hausenloy et al.

Blood pressure oscillations during tilt testing

1699

blood pressure trace with variations in systolic blood pressure of ,30 mmHg (peak-to-trough) occurring during either the passive tilt phase or during the glyceryl trinitrate (GTN) provocation.

Blood pressure oscillation as a predictive marker of vasovagal syncope Therefore, the presence of characteristic oscillations in blood pressure of 30 mmHg during HUT as a predictive marker for vasovagal syncope has a sensitivity of 88% (a positive predictive value of 87%) and a specificity of 40% (a negative predictive value of 43%) (Figure 3).

Discussion The main findings of this study are: (i) the presence of haemodynamic instability as indicated by oscillations in blood pressure, such that the systolic blood pressure varies 30 mmHg (peak-to-trough), strongly predicts a positive tilt test; (ii) the presence of such characteristic oscillations in blood pressure in a patient in which passive HUT and GTN provocation have not induced syncope, strongly predicts a positive test with Iso provocation, thereby possibly obviating the need for Iso testing in this patient group, and (iii) a significant proportion of patients non-responsive to passive HUT and GTN provocation may still have vasovagal syncope as diagnosed by Iso testing. Importantly, the presence of this oscillatory blood pressure response had a sensitivity of 88% and positive predictive accuracy of 87% for subsequent vasovagal syndrome. In contrast, its specificity was only 40% but its negative predictive accuracy was 43% for a subsequent negative test. These results strongly suggest that the presence of haemodynamic instability in patients undergoing

passive HUT and GTN provocation, in which symptoms of syncope have not been reproduced, can be diagnosed as having vasovagal syncope. Conversely, in patients who do not respond to either HUT or GTN provocation, the absence of this haemodynamic response during passive HUT and GTN provocation can exclude a diagnosis of vasovagal syncope as revealed by subsequent Iso testing. However, there exists a group of patients (19.5%) that respond to HUT and GTN provocation, who do not display the typical oscillating blood pressure response, suggesting that the presence of this haemodynamic response as a diagnostic test should be reserved for those non-responsive to HUT and GTN, in whom the diagnosis of vasovagal syncope is still suspected. Most other centres which investigate syncope according to the current ESC Guidelines1 use either GTN or Iso as a provocative agent for inducing syncope. In our centre, those patients nonresponsive to passive HUT and GTN provocation, in which the diagnosis is still suspected based on history and blood pressure oscillations, go on to have Iso provocation. A large proportion of these patients are diagnosed as having vasovagal syncope with Iso provocation. The important message from this finding is that centres in which drug provocation is restricted to GTN alone may be failing to diagnose vasovagal syncope in an appreciable number of patients.11 The pathophysiology underlying vasovagal syncope is not fully understood. However, the results provided by this study suggest that haemodynamic instability, as indicated by oscillations in

Downloaded from by guest on November 15, 2015

Figure 2 Representative chart recording of a typical oscillation-negative patient demonstrating a non-oscillating relatively stable systemic

1700

D.J. Hausenloy et al.

systemic blood pressure (30 mmHg peak-to-trough), may be a manifestation of a failing auto-regulatory mechanism that may contribute to the mechanisms underlying vasovagal syncope. The cause of this blood pressure variability is unclear, although it has been attributed to attenuated baroreceptor sensitivity,8,9 and possible variations in sympathetic activity occurring on a background of low parasympathetic activity,8 resulting in an impaired ability to buffer the blood pressure oscillations. Study limitations include the fact that this is a single centre retrospective study of a relatively small number of patients in a case control format. To summarize, the oscillation of systolic blood pressure during tilt testing points to a diagnosis of vasovagal syncope. Beat-to-beat blood pressure monitoring opens greater possibilities of diagnosis on tilt. It may not be necessary to pursue tilt tests to the point of collapse.12 The findings of this study point to future possibilities of monitoring for blood pressure oscillations with an external device to predict syncope possibly obviating the need to pursue a test to completion with unpleasant symptoms. Furthermore, it may be possible to detect these pre-syncopal blood pressure oscillations with an implanted device13 which then alerts the patient, enabling adoption of corrective measures (such as Isometric manoeuvres or lying down)14 or to self-administer medication to avert the syncopal episode.15 Conflict of interest: none declared.

Acknowledgements We would like to thank the staff and patients of the Chelsea and Westminster Hospital.

Funding We would like to thank the British Heart Foundation for their continuing support.

References 1. Brignole M, Alboni P, Benditt DG, Bergfeldt L, Blanc JJ, Bloch Thomsen PE et al. Guidelines on management (diagnosis and treatment) of syncope—update 2004. Europace 2004;6:467–537. 2. Bartoletti A, Alboni P, Ammirati F, Brignole M, Del Rosso A, Foglia MG et al. ’The Italian Protocol’: a simplified head-up tilt testing potentiated with oral nitroglycerin to assess patients with unexplained syncope. Europace 2000;2:339 –42. 3. Morillo CA, Klein GJ, Zandri S, Yee R. Diagnostic accuracy of a low-dose isoproterenol head-up tilt protocol. Am Heart J 1995;129:901 –6. 4. Natale A, Sra J, Akhtar M, Kusmirek L, Tomassoni G, Leonelli F et al. Use of sublingual nitroglycerin during head-up tilt-table testing in patients .60 years of age. Am J Cardiol 1998;82:1210 –3. 5. Zeng C, Zhu Z, Hu W, Liu G, Zhu S, Zhou Y et al. Value of sublingual isosorbide dinitrate before isoproterenol tilt test for diagnosis of neurally mediated syncope. Am J Cardiol 1999;83:1059 –63. 6. Sutton R, Bloomfield DM. Indications, methodology, and classification of results of tilt-table testing. Am J Cardiol 1999;84:10Q –19Q. 7. Brignole M, Menozzi C, Del Rosso A, Costa S, Gaggioli G, Bottoni N et al. New classification of haemodynamics of vasovagal syncope: beyond the VASIS classification. Analysis of the pre-syncopal phase of the tilt test without and with nitroglycerin challenge. Vasovagal Syncope International Study. Europace 2000;2: 66 –76. 8. Julu PO, Cooper VL, Hansen S, Hainsworth R. Cardiovascular regulation in the period preceding vasovagal syncope in conscious humans. J Physiol 2003;549: 299 –311. 9. Samniah N, Sakaguchi S, Ermis C, Lurie KG, Benditt DG. Transient modification of baroreceptor response during tilt-induced vasovagal syncope. Europace 2004;6: 48 –54. 10. Petersen ME, Williams TR, Sutton R. A comparison of non-invasive continuous finger blood pressure measurement (Finapres) with intraarterial pressure during prolonged head-up tilt. Eur Heart J 1995;16: 1641 –54.

Downloaded from by guest on November 15, 2015

Figure 3 Flow chart showing that of 83 patients with suspected vasovagal syndrome (VVS) recruited into the study, 42 patients had a negative head-up tilt (HUT) and went onto to receive isoprenaline testing, of which 27 patients were classified as positive and 15 were negative. The remaining 41 patients experienced a positive HUT and of these, 33 patients were classified as oscillation-positive (characteristic oscillating systemic blood pressure trace with variations in systolic blood pressure of 30 mmHg, peak-to-trough) and 8 patients as oscillation-negative (nonoscillating systemic blood pressure trace with variations in systolic blood pressure of ,30 mmHg, peak-to-trough). All the 27 isoprenaline positive patients were classified as oscillation-positive with no patients being oscillation-negative. Of the 15 isoprenaline negative patients, 9 of them were classified as oscillation-positive with 6 patients being oscillation-negative.

Blood pressure oscillations during tilt testing

11. Raviele A, Giada F, Brignole M, Menozzi C, Marangoni E, Manzillo GF et al. Comparison of diagnostic accuracy of sublingual nitroglycerin test and low-dose isoproterenol test in patients with unexplained syncope. Am J Cardiol 2000;85: 1194– 8. 12. Virag N, Sutton R, Vetter R, Markowitz T, Erickson M. Prediction of vasovagal syncope from heart rate and blood pressure trend and variability: experience in 1,155 patients. Heart Rhythm 2007;4:1375 –82. 13. Brignole M, Sutton R, Menozzi C, Garcia-Civera R, Moya A, Wieling W et al. Early application of an implantable loop recorder allows effective specific therapy in

1701 patients with recurrent suspected neurally mediated syncope. Eur Heart J 2006;27:1085 – 92. 14. Krediet CT, van Dijk N, Linzer M, van Lieshout JJ, Wieling W. Management of vasovagal syncope: controlling or aborting faints by leg crossing and muscle tensing. Circulation 2002;106:1684 – 9. 15. Raviele A, Giada F, Gasparini G. Efficacy of a patient-activated pharmacologic pump using phenylephrine as active drug and prodromal symptoms as a marker of imminent loss of consciousness to abort tilt-induced syncope. J Am Coll Cardiol 2005;45:320 –1.

Downloaded from by guest on November 15, 2015