ABSTRACT
Purpose Autonomic dysreflexia, often accompanied by heart rate changes, increases the risk of cardio-cerebrovascular complications in individuals with spinal cord injury. Thus, our aim was to characterize these changes during penile vibrostimulation and urodynamics.
Materials and Methods We analyzed the cardiovascular (i.e. blood pressure and heart rate) data from two prospective studies, i.e. 21 individuals with chronic spinal cord injuries and history of autonomic dysreflexia, who underwent penile vibrostimulation (n=11, study 1) or urodynamics (n=10, study 2).
Results The cohort’s median age was 41 years (range 22 −53). Overall 47 episodes of autonomic dysreflexia were recorded (i.e. penile vibrostimulation n=37, urodynamics n=10), while at least one episode was recorded in each participant. At the threshold of autonomic dysreflexia, bradycardia was observed during penile vibrostimulation and urodynamics in 43% and 30% of all episodes, respectively. At the peak of autonomic dysreflexia during penile vibrostimulation and urodynamics, bradycardia was observed in 65% and 50%, respectively. In contrast, tachycardia was detected only once during urodynamics.
Conclusion Our findings reveal that heart rate changes associated with autonomic dysreflexia during penile vibrostimulation and urodynamics appear to be related to the magnitude of systolic blood pressure increases. Thus, highly elevated systolic blood pressure associated with bradycardia suggest the presence of severe autonomic dysreflexia, which can lead to devastating cerebro-cardiovascular consequences. Therefore, we recommend cardiovascular monitoring during penile vibrostimulation and urodynamics to detect autonomic dysreflexia and stop assessments before systolic blood pressure is dangerously increasing, thereby reducing the risk of potentially life-threatening complications in this cohort.
INTRODUCTION
Long-term consequences of SCI are devastating and extensive.1 Of note, autonomic functions, including cardiovascular, LUT and sexual function are often impaired post-injury, thereby placing a tremendous burden on this population.2,3 AD is a well-known autonomic consequence following SCI4, which, if not treated, can lead to life-threatening complications including stroke, myocardial infarction or death.5
In daily life, AD can occur in response to intrinsic stimuli (e.g., bladder distention and UTI) and/or extrinsic stimuli (e.g., restrictive clothing).6 Although highly informative, commonly performed urological diagnostic procedures (e.g., PVS, UDS, and cystoscopy) can trigger severe AD episodes67,8 and associated cardiac arrhythmias.9 PVS is a well-established technique for sperm retrieval in males who wish to father children.10,11 However, due to its intense stimulation, PVS is associated with high cardiovascular risk.12 UDS, the gold standard to assess ANLUTD in SCI, is also well-known to elicit AD.13,14
Currently, HR responses associated with AD during both procedures are indeterminate. Although most studies reported HR decreases (i.e., relative bradycardia, i.e. <60bpm,5,8,9,15–17) as the primary response during AD, which is in line with our understanding of the pathophysiologic HR and BP changes18,19 (i.e., a parasympathetically-mediated HR decrease in order to lower SBP during AD 2,7,9,20–22), Solinsky et al. reported that relative tachycardia (i.e., >100bpm), not bradycardia (i.e. 68% vs. 0.3%) was the primary HR response during AD.23 Given this conflicting literature, we aimed to characterize iatrogenically induced AD (i.e. during urodynamics or sperm retrieval) associated HR changes during both procedures using continuous beat-to-beat BP monitoring, a monitoring system crucial in order to capture peak systolic change.12 Considering the lack of official safety protocols and the evidence of well-known cardio-cerebrovascular complications associated with AD, better knowledge of cardiovascular changes associated with AD (e.g., HR changes) is essential to improve the safety of individuals with SCI.
MATERIALS AND METHODS
Study design and ethics
This is an analysis of complete cardiovascular data sets from two prospective studies (i.e., one study not qualifying as an applicable clinical trial, and one clinical trial – NCT02676154), approved by the University of British Columbia Clinical Research Ethics Board, Vancouver Coastal Health Research Institute and Health Canada.
Participants
We screened for complete records of individuals with chronic SCI and NLI at or above the sixth thoracic spinal level who underwent PVS or UDS. Complete inclusion and exclusion criteria, study design and diagnostic assessments are described in the supplementary methods.
Outcome
Outcome was the characterization of HR changes (i.e. increase, decrease, tachycardia or bradycardia) at AD threshold (i.e., change [Δ] in SBP >20mmHg from baseline) and AD peak (i.e. maximum ΔSBP during AD). In accordance with the ISAFSCI, AD was defined as an increase in SBP of >20 mmHg from baseline.6
Hemodynamic Assessments
Cardiovascular parameters were recorded continuously for ten minutes before (i.e., baseline) and during the UDS and PVS. HR (ECG, lead II; Powerlab Model ML132, ADInstruments, Colorado Springs, Colorado) and beat-to-beat BP (via photoplethysmography; Finometer, Finapres Medical Systems BV, Arnhem, The Netherlands) were monitored non-invasively. Data were collected at a rate of 1kHz per channel through an analog to digital converter (Powerlab/16SBP model ML795, ADInstruments, Colorado Springs, Colorado) and calibrated to brachial BP measurements (Dinamap 100, GE, Fairfield, Connecticut, USA) taken in 1-minute intervals throughout each procedure. The time to and the duration of AD were recorded during PVS. Continuous data collection occurred throughout each procedure using LabChart® software (Version 8.0, ADInstruments, Colorado Springs, Colorado). All figures were created with Prism Version 8.4.0 (GraphPad Software, San Diego, California). The protocols, devices used, and data extraction, are explained in the supplementary methods.
Statistical Analysis
Statistical analyses were conducted using R Statistical Software Version 3.6.0 for Macintosh Operating System. Data are presented as raw values and percentage, and median with IQR. Range (i.e., min -max) is provided for age, time post-injury, and number of PVS per individual. In LabChart® software, raw data were extracted as a 3-beat average per point based on the ECG. From these averaged data points, the cardiovascular parameters from one minute prior to recording were calculated as the average (i.e., mean). For each AD episode during PVS and UDS, AD threshold and AD peak were identified. Then, all corresponding HRs were identified to calculate ΔHR from baseline, i.e., at AD threshold and AD peak. In PVS only, baseline parameters were compared to the relative response for each stimulation, and new baseline data were established for each stimulation. AD threshold for each stimulation was identified based on each re-calculated baseline. The median and IQR were calculated individually for each PVS, and all stimulations taken together. SBP and HR were also determined at the point of ejaculation for eligible participants. In addition, the median and IQR of the grouped data are reported. In UDS only, we analyzed documented clinical signs and symptoms.
RESULTS
Participants
In total, 21 individuals (median age 41 years [IQR 37–47, range 22–53], with a median post-injury time of 18 years [IQR 7–27, range 4–39]), who underwent PVS (11/21, all male) or UDS (10/21, including 4 females) with complete recordings were included in our analysis. Demographics and injury characteristics of participants are shown in table 1.
Cardiovascular parameters
Table 2 provides an overview of SBP and HR values at baseline, at AD threshold and at AD peak including their changes where appropriate.
Penile vibrostimulation
During an average of 3 PVS per participant (range 1 to 5), we identified 37 AD episodes across all individuals. Figure 1 depicts the individual ΔSBP during each PVS (data shown in Table 2) as well as HR changes at AD threshold and peak AD. At baseline, median SBP and HR were 114mmHg (104 – 120) and 65bpm (59 – 73) for all 37 AD episodes. Upon reaching AD threshold, the median HR was 62bpm (56 – 76) with a median ΔHR of 1bpm (−5 – 5). At AD threshold, HR increased in 51% (19/37, median Δ = 5bpm, 2 – 8), decreased in 41% (15/37, median Δ = −7bpm, −13 – −3) or remained unchanged in 8% (3/37). Furthermore, bradycardia was present in 43% episodes (16/37, median HR 55bpm, 52 – 57), while in 57% (21/37) median HR was 76bpm (64 – 80). At AD peak (median ΔSBP 67mmHg, 51 – 99), an associated median HR (−10bpm, −17 – −5) was recorded across all participants. As well, the majority of HR changes were classified as bradycardia (65%, 24/37, median 51bpm, 43-54), while the remaining 35% (13/37) HR changes remained below the threshold for tachycardia (median 72bpm, 69 – 73). The moment of ejaculation during PVS was captured in 4 individuals, revealing a median ΔSBP of 98mmHg (84 – 123) at AD peak with an associated median ΔHR of −19bpm (−23 – −17).
Urodynamics
During a total of 10 UDS, we identified an episode of AD in each individual (all motor-complete SCI, i.e. AIS A/B). Figure 2 depicts the individual ΔSBP during each UDS (data shown in Table 2) as well as HR changes at AD threshold and peak. At AD threshold, a median HR of 65bpm (55 – 83) and median ΔHR of 0bpm (−3 – 2) was see across all participants. Furthermore, HR increased in 40% (4/10, median Δ = 3bpm, 3 – 7), decreased in 40% (4/10, median Δ = −4bpm, −6 – −3) or remained unchanged in 20% (2/10). Bradycardia was observed for 30% of episodes (3/10, median HR = 53bpm, 52 – 53), and the remaining 70% (7/10) of episodes showed a median HR of 79bpm (71 – 81). At AD peak (median ΔSBP 53mmHg, 43–75), an associated median HR (−1bpm, −8 – 1) was recorded. In 7 individuals an AD-associated HR decrease was observed. Five individuals (median = 53bpm, IQR = 48-55) experienced bradycardia, while 4/10 had a median HR of 73bpm (66 – 75). Only one individual (T3 AIS A), experienced tachycardia (102bpm) at AD peak. Table 3 summarizes the signs and symptoms of AD during the UDS.
DISCUSSION
Main findings
The primary HR response at AD peak (i.e., compared to baseline) during PVS and UDS was a decrease in HR. Bradycardia, not tachycardia was mainly observed during PVS and in half the time during UDS. Only one participant had a tachycardia at AD peak during UDS. This pattern of AD associated HR change during both assessments appears to be dependent on the ΔSBP.
Findings in context of existing evidence
The current findings are in line with the previous understanding of a parasympathetically-mediated response to counterbalance extensive SBP increases during AD.15 Considering the impairment of bulbospinal pathways following SCI, inhibitory efferent signals cannot reach and neutralize the activated SPNs below the NLI.24 Thus, vasoconstriction and thereby increased SBP are maintained as long as the AD stimulus is activated. Consequently, baroreceptors are activated, resulting in a decrease of HR, which can result in bradycardia and even asystole.25,26 Several research groups have observed and reported this pattern.8,9,15,16,22 Our observation of HR decreases at the peak of AD displays an impaired supraspinal control of the heart and SPNs below the NLI, accompanied by unopposed parasympathetical (i.e. vagal) tone during AD.9 In this current study, SBP reached as high as 300mmHg during PVS, potentially revealing the SBP-dependent effect on HR that was not shown in the 2018 study.23
The current study purposefully utilized continuous beat-to-beat BP monitoring to ensure greater accuracy of SBP peaks during AD episodes12, compared to intermittent BP monitoring. Continuous BP monitoring increases the validity of our findings by accurately detecting rapid and extreme SBP fluctuations.
Additionally, these HR findings are contrary to those from Solinsky et al., who reported relative tachycardia as the primary HR response during AD.23 This 2018 study was different from the design of our current study in that it had a more conservative definition of AD and no specific interventions. Our study measured SBP specifically during highly stimulating interventions. Furthermore, the Solinsky study measured BP manually and intermittently versus continuously, reducing the accuracy of the result and possibly underestimating the true SBP value.
However, this secondary analysis was neither powered nor designed to investigate HR changes specific to the NLI or severity (i.e. AIS grading). Nevertheless, we showed in a previous study that not only the higher the NLI, the greater the odds are to experience AD during UDS but those with a complete (AIS A) vs. incomplete SCI (AIS B – D) had greater AD-associated HR decreases.7
When considering the completeness of injury (i.e., AIS grade) and its impact on ΔHR and ΔSBP (i.e., AD severity), it is suggested that incomplete injuries will show better preserved cardiovascular function through reduced incidence of AD and increased low-frequency and high-frequency HR variability at rest.27 In both, UDS and PVS analyses there was high variability in ΔSBP and ΔHR response when considering completeness and level of injury. Our studies were not aimed at investigating HR changes specific to completeness of NLI, but instead to provide first insights to be thoroughly investigated in a future prospective trial.
In the UDS study only, AD symptoms were documented in all participants. This was done specifically in support of the findings of a 2017 study, showing that a patients age (>45 years), completeness of injury, and the symptoms of chills or sweating, are risk factors for AD symptoms occurring during UDS.28 In our study all participants had a complete SCI (i.e., AIS A or B), half the cohort was over 45 years of age, and 30% presented with sweating during UDS. Risk factor prediction of AD symptoms is a potentially useful tool in the clinical setting especially as patients are able to state their symptoms throughout a procedure. Additionally, a 2018 study suggested that neurogenic detrusor overactivity was also a predictor for AD occurring during UDS.7 The current UDS study supports these findings, where all participants exhibited neurogenic detrusor overactivity and AD during the procedure.
Implications
These humble, novel findings might help inform the management of individuals with AD undergoing PVS and UDS, potentially other diagnostic procedures that could provoke AD. Although beat-to-beat cardiovascular monitoring is considered ideal, these devices are mainly found in research institutes. However, monitoring SBP every minute, as a minimal requirement for safety surveillance with regard to AD, seems feasible in a busy daily clinic. Since AD can be severe and asymptomatic in individuals undergoing PVS22,29 and UDS7, these findings may also aid to an increased awareness that rapid SBP increases can occur without clinical signs and symptoms, and without SBP monitoring would remain undetected and potentially lead to devastating cardiovascular consequences. Additionally, monitoring HR along with SBP during these procedures may help to recognize a profound and potentially dangerous episode of AD and thereby assist in knowing when to implement a safe stop protocol. Clinical signs and symptoms, including patient age, NLI and personal characteristics for prediction of AD can be utilized in a similar way to encourage a safer outcome.
Future research should focus on elucidating the mechanisms behind these HR responses, in addition to obtaining a wider range of data from different NLI or AIS grade. Future studies applying this knowledge in a clinical environment will be helpful for addressing its practical applications.
Study limitations
Our study was limited by a small cohort of participants and the distribution of sex and injury characteristics. The small sample size limits the conclusion drawn from the NLI and its effect on HR outcomes in AD. The majority of participants were males (81%, 17/21), limiting the findings of our female cohort. Except for one participant with a motor -incomplete SCI (i.e. AIS C), all participants had a motor -complete SCI (AIS A / B), limiting the application of our findings solely to the latter group. Furthermore, our study participants all had chronic injuries, so we cannot comment on the characteristics of HR responses during PVS and UDS in patients with acute SCI. Additionally, there were four methods used during the PVS procedures, each with different modes and intensities, with the goal of maximizing the occurrence of ejaculation, which may have influenced the degree of AD experienced by participants in the PVS study. Given these limitations, further studies that include a greater variety of levels and completeness of injury would help to inform and individualize the HR response in each type of injury.
CONCLUSION
This secondary analysis was aimed at characterizing the HR responses throughout AD to the point of maximum SBP increase (i.e. peak AD) in two common diagnostic procedures applied in daily urological practice potentially eliciting an AD episode. The findings document clinical bradycardia during AD as the primary response during the peak of AD in individuals with a motor-complete SCI, creating new implications for the management of AD during PVS and UDS. Although cardiovascular monitoring during these assessments is already recommended to reduce the risk of life-threatening complications of AD5, there is a potentially novel role for ΔHR to help characterize and identify peak AD. Considering that AD severity during UDS and PVS is supported by a decrease in HR, which seems to culminate at maximum ΔSBP, monitoring HR along with SBP may help identify the development of a serious AD episode in this cohort.
Data Availability
The data are available from the corresponding author on reasonable request.
Author contributions
Andrei V. Krassioukov and Matthias Walter had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design
Lauren Rietchel, Andrea L. Ramirez, Stacy Elliott, Andrei V. Krassioukov, Matthias Walter
Acquisition of data
Andrea L. Ramirez, Shea Hocaloski, Stacy Elliott, Andrei V. Krassioukov, Matthias Walter
Analysis and interpretation of data
Lauren Rietchel, Andrea L. Ramirez, Shea Hocaloski, Stacy Elliott, Andrei V. Krassioukov, Matthias Walter
Drafting of the manuscript
Lauren Rietchel
Critical revision of the manuscript for important intellectual content
Lauren Rietchel, Andrea L. Ramirez, Shea Hocaloski, Stacy Elliott, Andrei V. Krassioukov, Matthias Walter
Statistical analysis
Lauren Rietchel and Matthias Walter
Supervision
Andrei V. Krassioukov and Matthias Walter
Author responsibilities
According to the ICJME authorship, all authors provided substantial contributions to the conception or design of the work, or the acquisition, analysis or interpretation of data for the work; AND drafting the work or revising it critically for important intellectual content; AND final approval of the version to be published; AND agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Obtaining funding
This study was funded by the Rick Hansen Man in Motion Research Foundation (grant number: 135774, awarded to Dr. Krassioukov). The research equipment for this study was supported by the Canadian Foundation of Innovation (CFI, grant number: 35869) and the British Columbia Knowledge Development Fund (BCKDF, grant number: 35869. Dr. Walter was supported by a 2017-2019 Michael Smith Foundation for Health Research (MSFHR) and Rick Hansen Foundation Postdoctoral Research Trainee Award (grant number 17110). Dr. Krassioukov holds the Endowed Chair in Rehabilitation Medicine.
Competing interests
The authors have no competing interests to disclose.
Acknowledgements
We wish to acknowledge the Blusson Spinal Cord Centre. Furthermore, we would like to thank the participants who provided written informed consent for participation and publication in this study.
Footnotes
↵# shared senior authorship
ABBREVIATIONS
- AD
- autonomic dysreflexia
- AIS
- American Spinal Injury Association Impairment Scale
- ANLUTD
- adult neurogenic lower urinary tract dysfunction
- BP
- blood pressure
- ECG
- electrocardiogram
- HR
- heart rate
- IQR
- interquartile range
- ISAFSCI
- International standards to document remaining autonomic function after Spinal cord injury
- LUT
- lower urinary tract
- NLI
- neurological level of injury
- PVS
- penile vibrostimulation
- QOL
- quality of life
- SBP
- systolic blood pressure
- SCI
- spinal cord injury
- SPN
- sympathetic preganglionic neuron
- UDS
- urodynamic studies
- UTI
- urinary tract infection