Short-duration head-up tilt test potentiated with sublingual nitroglycerin in suspected vasovagal syncope: the fast Italian protocol

Abstract

Background

The traditional nitroglycerin (NTG) head-up tilt test (HUTT) is time-consuming and the test duration is a barrier to widespread utilization in clinical practice. It was hypothesized that a short-duration protocol is not inferior to the traditional protocol regarding the positivity rate and has a similar distribution of hemodynamic response.

Methods and results

Patients undergoing HUTT were randomized 1:1 to a 10 min passive phase plus a 10 min 0.3 mg NTG if the passive phase was negative (Fast) or to a 20 min passive phase plus a 15 min 0.3 mg NTG if the passive phase was negative (Traditional). A sample size of 277 patients for each group achieved 80% power to detect an expected difference of 0% with a non-inferiority margin of -10% using a one-sided t-test and assuming a significant level alpha of 0.025. A total of 554 consecutive patients (mean age 46.6 ± 19.3 years, 47.6% males) undergoing HUTT for suspected vasovagal syncope were randomly assigned to the Fast (n = 277) or Traditional (n = 277) protocol. A positive response was defined as the induction of syncope in presence of hypotension/bradycardia, and was observed in 167 (60.3%) patients with Fast and in 162 (58.5%) patients with the Traditional protocol. There was a trend of lesser vasodepressor response (14.8% Fast vs. 20.6% Traditional) which was significant during the passive phase (P = 0.01).

Conclusion

The diagnostic value of the Fast HUTT protocol is similar to that of the Traditional protocol and therefore the Fast protocol can be used instead of the Traditional protocol.

Structured Graphical Abstract

Fast Italian protocol: positivity rate and type of haemodynamic responses compare to traditional HUTT protocol.

Open in new tabDownload slide

See the editorial comment for this article ‘Tilt testing evolves: faster and still accurate’, by Artur Fedorowski et al., https://doi.org/10.1093/eurheartj/ehad359.

Introduction

Head-up tilt test (HUTT) is a useful diagnostic tool for patients with suspected vasovagal syncope (VVS) after initial clinical assessment.¹ Owing to its high diagnostic yield with a sensitivity of 66% and a specificity of 89%,² ease to perform without the need for an intravenous line, and safety virtually without complications, the nitroglycerin (NTG) potentiated protocol (the so-called ‘Italian protocol’³) is probably the most widely used HUTT in clinical practice, and it is recommended by the 2018 European Society of Cardiology (ESC) guidelines.⁴ Nevertheless, it is time-consuming, and its duration is one of the barriers to widespread utilization in clinical practice. It was hypothesized that a short-duration protocol is not inferior to the traditional protocol in regards to the positivity rate and has a similar distribution of hemodynamic response.

Methods

Consecutive patients undergoing HUTT for diagnosis of suspected VVS at the Syncope Units of Naples and Bolzano hospitals, Italy, from 1 July 2021 to 3 September 2022, were randomized 1:1 with a centralized non-stratified block procedure to the short (Fast) or the usual (Traditional) NTG potentiated HUTT. The exclusion criteria were: (i) non-syncopal causes of real or apparent loss of consciousness that may be incorrectly diagnosed as syncope (e.g. unexplained falls, epilepsy, psychogenic pseudosyncope and other rare causes); (ii) established or suspected cardiac syncope in accordance with the criteria of the ESC syncope guidelines⁴; (iii) classical orthostatic hypotension and postural orthostatic tachycardia syndrome diagnosed at the initial evaluation by the active standing test.

The Fast Italian protocol included a tilt passive phase of 10 min at 70°, and, in case of negativity, a sublingual NTG spray administration, at a fixed dose of 300 µg, followed by an active phase of 10 min. In the Traditional Italian protocol, the duration of passive and NTG phases of tilt was 20 and 15 min respectively, at 70°, as previously described.^3,4 Both Fast and Traditional protocols were preceded by a supine pre-tilt phase of 5 min without venous cannulation. The HUTT continued until complete loss of consciousness occurred, indicated by the lack of response to a vocal stimuli, loss of muscle tonus, and jerking movements or the completion of the protocol without syncope, whichever occurred first. Reproduction of spontaneous symptoms was not used as a criterium of HUTT positivity. For this study, in the case of development of progressive hypotension during the test (i.e. delayed orthostatic hypotension) not followed by syncope, the test was considered negative.

If a vasovagal pattern of hypotension/bradycardia or symptoms of impending syncope were present at the end of the passive phase or of the NTG phase, that phase was continued until syncope had occurred or for a maximum of 3 extra minutes. The time duration to tilt down the motorized table was 12 s. The HUTT responses were classified according to the new VASIS classification.⁵ The test was performed during continuous electrocardiogram (ECG) and blood pressure (BP) monitoring using the TaskForce monitor (CNSystem, Graz, Austria). This study was conducted according to the Declaration of Helsinki and approved by the institutional ethics committees (ID-168/02032021); written informed consent was obtained from the patients.

Statistical analysis

A sample size of 277 patients for each group achieved 80% power to detect an expected difference of 0% with a non-inferiority margin of -10% using a one-sided t-test and assuming a significant level alpha of 0.025. The normal distribution of the data was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. Normally distributed variables were expressed as mean ± standard deviation. Categorical variables were reported as numbers and percentages. Normally distributed continuous variables were compared using Student’s t-test for paired and unpaired comparisons, as appropriate; categorical variables were compared with the Fisher exact test. An α level of 0.05 with Bonferroni correction for the number of hypotheses was considered significant for all measurements. All statistical analyses were performed using STATA software version 11.1SE (StataCorp, College Station, TX, USA).

Results

Population

A total of 554 consecutive patients (mean age 46.6 ± 19.3 years, 47.6% males) with suspected VVS after the initial clinical evaluation were randomly assigned to Fast (n = 277) or Traditional (n = 277) HUTT protocol. The initial clinical evaluation included careful history taking concerning present and previous attacks, physical examination, supine and standing BP measurements, and ECG; moreover, ECG monitoring, echocardiogram, and carotid sinus massage were performed when indicated. The patient’s clinical characteristics were well-balanced in the two groups (Table 1).

Comparison between groups

A HUTT positive response was observed in 167 (60.3%) patients of the Fast group and 162 (58.5%) patients of the Traditional group, thus resulting in +2.8% absolute positivity rate of the HUTT Fast protocol, thus meeting the criteria for non-inferiority (Table 2).

In the Fast group, there was a trend of lower prevalence of vasodepressor response than in the Traditional group (14.8% vs. 20.6%), which was significant during the passive phase (P = 0.01). The distribution of different HUTT responses during passive and NTG-potentiated phases is shown in Table 2.

The cumulative number of patients with a positive HUTT response (by 1 min time frame) during both passive and NTG phases in the two groups is shown in Figure 1. The slope of the curve was greater around the 10th minute in both groups; the slope was almost stable from the 11th to the 20th minute in the Traditional group.

During the passive phase, 16 patients in the Fast group and 23 patients in the Traditional group had syncope; further three patients in the Traditional group, who had a vasovagal pattern of hypotension/bradycardia or symptoms of impending syncope at the 20th minute, developed syncope within the next 3 min. Thus, a total of 16 (5.8%) and 26 (9.4%) patients had syncope during the passive phase (P = 0.07). Furthermore, 22 patients showed a pattern of delayed orthostatic hypotension during the passive phase: 14 in the Traditional group and 8 in the Fast group (P = 0.19). All these patients developed syncope during the subsequent NTG phase (mixed form in 8 and vasodepressor form in 14) and were therefore counted as positive responses of the NTG phase.

During the NTG phase, 148 patients in the Fast group and 135 patients in the Traditional group had syncope; further three patients of the Fast and one patient of the traditional group, who had a vasovagal pattern of hypotension/bradycardia or symptoms of impending syncope, were maintained in the tilt-up position and developed syncope within the next 3 min. Thus, a total of 151 (54.5%) and 136 (49.1%) patients had syncope during the NTG phase (P = 0.23).

Blood pressure and heart rate values measured at the end of the supine phase and the 1st, 10th, and 20th minute of the passive phase and the first minute of the NTG phase are reported in the Table 3. In the Traditional group, systolic BP initially decreased from 127.4 ± 16.8 mmHg to 123.1 ± 16.6 mmHg during the 10 min passive phase, and then increased to 126.8 ± 16.7 mmHg at the end of the passive phase (P = 0.0006 compared to 10th min); finally, it decreased to 112.9 ± 16.1 mmHg at 1 min after NTG administration (Figure 2). In the Fast group, systolic BP decreased from 128.5 ± 17.4 mmHg to 122.5 ± 17.0 mmHg during the passive phase and to 111.4 ± 13.4 mmHg at 1 min after NTG administration (Figure 2). The difference in systolic BP at the end of the passive phase between the two groups was 4.3 mmHg (P = 0.003). The changes in diastolic BP were concordant with the above pattern of systolic BP. Heart rate progressively increased.

Discussion

The main result of this study is that the positivity rate of the Fast HUTT protocol is non-inferior to that of the Traditional protocol for diagnosing VVS, despite HUTT duration is reduced by 38% from 40 to 25 min. Overall, the saved time was of 136 h. Moreover, the Fast protocol did not significantly influence the type of haemodynamic responses except for a slightly lower prevalence of vasodepressor response during the passive phase (Structured Graphical Abstract). This finding is consistent with the data shown by van Dijk et al.⁶ that vasodepression takes 8 min to develop and does not necessarily start immediately after tilt-up.

The evidence that the reduction of HUTT duration does not impact on its positivity rate requires an explanation. Several factors have contributed to this finding:

• Passive phase duration. The contribution of the 20 min passive phase to HUTT positivity was modest, accounting for 9.4% in our population and 11% in the previously published data.³ In the Traditional group, 5.4% of patients had syncope within the first 10 min, a figure similar to 5.8% observed in the Fast group. Thus, the shortening of the passive phase to 10 min determined a loss of 4% of positive cases, mainly those with vasodepressor form or delayed orthostatic hypotension, which was compensated by the higher positivity of the NTG phase in the Fast group (see below). This finding is consistent with the literature. Indeed, in the meta-analysis of Forleo et al.,² which included 55 studies for a total of 4361 patients, a longer passive phase increased the positivity rate of the unmedicated phase while reducing the proportion of patients fainting during the pharmacological phase; therefore, the total HUTT positivity rate was not influenced by the different duration of the passive phase. Conversely, a passive phase duration shorter than 10 min seems to be unfavorable, since, in a randomized study,⁷ the positivity rate of an accelerated NTG potentiated HUTT, consisting of a 5 min passive plus a 20 min NTG phase, was inferior to that of the conventional protocol (51% vs. 35%, P = 0.04). Thus a 10 min passive phase is necessary to optimize the positivity rate of HUTT. Furthermore, 10 min of passive tilting are sufficient to diagnose patients affected by delayed orthostatic hypotension.

• Blood pressure changes during the passive phase. The optimal 10 min duration of the passive phase is consistent with the pattern of systolic BP changes observed during this phase (Figure 2). In both the Fast and Traditional groups, we observed a slight decrease in the systolic BP from starting of the upright position until the 10th minute, with similar values between the groups. In the Fast group, systolic BP continued to decrease as a consequence of NTG administration; conversely, in the Traditional group, systolic BP increased from the 10th to the 20th minute. Thus, the difference between groups in systolic BP at the end of passive phases was 4.3 mmHg (P = 0.003). Our interpretation of this original finding is that the effect of orthostatic stress was maximum at the 10th minute, and after that the systolic BP increased as a consequence of the activation of the compensatory hemodynamic mechanisms (i.e. baroreceptor activation and neuroendocrine response).^6,8 As a consequence, the hemodynamic effect of the NTG administration was more powerful and rapid when it was administered in an already predisposed critical situation, as it happens in the Fast group; conversely, it was less powerful and slow when it was administered in a more stable hemodynamic condition, as observed in the Traditional group.

• NTG phase duration. In the original description of the Italian protocol,³ a NTG phase duration of 15 min was recommended (if NTG spray administration was adopted instead of NTG tablet), because the mean time to syncope was 5 ± 4 min. Glockler et al.⁹ have calculated that shortening the active phase from 15 to 10 min would result in a positivity rate loss of 1.5%. The boost effect of NTG administration at the 10th minute of the passive phase was able to speed up and facilitate the vasovagal reflex in the Fast group which showed a shorter time to syncope. Indeed, we showed a significantly increased HUTT positivity during the 10 min NTG phase in the Fast group compared to the first 10 min of NTG phase in the Traditional group (151 vs. 101 patients, P = 0.0001). As a consequence, the duration of the NTG phase could be reduced from 15 min in the Traditional to 10 min in the Fast protocol.

Practical implications

HUTT is time-consuming, and its long duration is one of the barriers for widespread utilization in clinical practice. Owing to its long duration, the test is expensive and often not fully reimbursed by the healthcare services. In western countries, the price of the test is several hundred euros even though it varies across countries and hospitals. For these reasons, the duration of the test has been historically and progressively reduced from the initial 60 min of passive only phase¹⁰ or from the initial 45 min of passive plus 20 min of NTG phase in the potentiated HUTT protocol.¹¹ The duration of the Traditional Italian protocol,³ although shorter than many earlier ones, is still long and a deterrent to its use. The Fast Italian protocol maintains the advantage of the Traditional protocol by avoiding the need for an intravenous line with its insertion possibly polluting the autonomic atmosphere as well as taking time and requiring a longer passive time at the outset. Further, sublingual NTG is inexpensive and benign, two facets not enjoyed by isoproterenol potentiation.

The total duration of the Fast protocol is 25 min (including the initial 5-min supine pre-tilt phase) and it is similar to that of other non-invasive provocative tests widely used in cardiology such as pharmacological stress echo, exercise testing, or stress scintigraphy. We are confident that the Fast Italian HUTT will be accepted by physicians who can save time and by health administrators who can reduce costs, thus finally favoring its widespread utilization in clinical practice.

Limitations

The specificity of the Fast HUTT protocol in a population without a history of syncope was not assessed. However, since in the literature the specificity remains relatively constant at a value of 10% or less, irrespective of NTG tilt protocol,^2,3 we can suppose that this figure may be applied also to the Fast protocol. Introducing NTG at the 10th min of the passive phase may potentially reduce the diagnostic sensibility for delayed orthostatic hypotension; however, among our study population, no significant difference in orthostatic hypotension diagnosis was shown between groups. Although the COVID-19 outbreak caused a reduction of ∼50% in the number of HUTT performed during the study period compared to historical data, we did not observe an impact on the case mix of the study population and therefore we do not expect an impact on the study results.

Conclusions

In patients with suspected VVS, the diagnostic value of the Fast Italian HUTT protocol is similar to that of the Traditional protocol, with no significant difference in the distribution of hemodynamic responses. Therefore, the Fast protocol can be used instead of the Traditional protocol allowing to reduce the test duration from 40 to 25 min.

Data availability

The data underlying this article will be shared on reasonable request to the corresponding author.

Funding

All authors declare no funding for this contribution.

References

Author notes