Autonomic dysreflexia (AD), often accompanied by heart rate (HR) changes, increases the risk of cardio-cerebrovascular complications, such as stroke, myocardial infarction or death in individuals with spinal cord injury (SCI). [1]
Although highly informative, commonly performed urological diagnostic procedures, e.g., urodynamics (UDS) or penile vibrostimulation (PVS), a well-established technique for sperm retrieval, can trigger severe AD episodes and associated cardiac arrhythmias. [2] 
Currently, HR responses associated with AD during both procedures are indeterminate. Although most studies reported HR decreases (i.e., relative bradycardia, i.e. <60bpm) as the primary response during AD, which is in line with our understanding of the pathophysiologic HR and blood pressure (BP) changes (i.e., a parasympathetically-mediated HR decrease in order to lower systolic BP during AD), 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. [3]
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. 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.

This is an analysis of complete cardiovascular data sets from two prospective studies, approved by the local ethics committee. We screened for complete records from individuals with chronic SCI and neurogenic level of injury (NLI) at or above the sixth thoracic spinal level who underwent PVS or UDS. 
Primary outcome measure was the characterization of HR changes (i.e. increase, decrease, tachycardia or bradycardia) at AD threshold (i.e., change in  systolic BP >20mmHg from baseline) and AD peak (i.e. maximum change in systolic BP during AD). In accordance with the International standards to document remaining autonomic function after SCI (ISAFSCI), AD was defined as an increase in systolic BP of >20 mmHg from baseline.
Cardiovascular parameters were recorded continuously for ten minutes before (i.e., baseline) and during the UDS and PVS. HR and beat-to-beat BP were monitored non-invasively. Data were collected at a rate of 1kHz per channel through an analog to digital converter and calibrated to brachial BP measurements taken in 1-minute intervals throughout each procedure. The time to and the duration of AD were recorded during PVS.
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 addition, the median and IQR of the grouped data are reported. In UDS only, we analyzed documented clinical signs and symptoms.

We included complete data sets from 21 individuals with chronic SCI and history of AD, who underwent PVS (n=11, study 1) or UDS (n=10, study 2). The cohort's median age was 41 years (IQR 37–47, range 22–53) with a median post-injury time of 18 years (IQR 7–27, range 4–39). Overall 47 episodes of AD were recorded (i.e. PVS n=37, UDS n=10), while at least one episode was recorded in each participant. 
During an average of 3 PVS per participant (range 1 to 5), we identified 37 AD episodes across all individuals. At baseline, median systolic BP 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 HR change of 1bpm (-5 – 5). At AD threshold, HR increased in 51% (19/37, median change = 5bpm, 2 – 8), decreased in 41% (15/37, median change = -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), i.e. remaining below the threshold for tachycardia. At AD peak (median systolic BP change 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 other 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 systolic BP change of 98mmHg (84 – 123) at AD peak with an associated median HR change of -19bpm (-23 – -17). 
During a total of 10 UDS, we identified an episode of AD in each individual (all motor-complete SCI, i.e. AIS A/B). At AD threshold, a median HR of 65bpm (55 – 83) and median HR change of 0bpm (-3 – 2) was see across all participants. Furthermore, HR increased in 40% (4/10, median change = 3bpm, 3 – 7), decreased in 40% (4/10, median change = -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 systolic BP change 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.

The current findings are in line with the previous understanding of a parasympathetically-mediated response to counterbalance extensive SBP increases during AD. 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. This pattern of AD associated HR change during both assessments appears to be dependent on the systolic BP change.

Our findings reveal that HR changes associated with AD during PVS and UDS appear to be related to the magnitude of systolic BP increases. Thus, highly elevated systolic BP associated with bradycardia suggest the presence of severe AD, which can lead to devastating cerebro-cardiovascular consequences. Therefore, we recommend cardiovascular monitoring during PVS and UDS to detect AD and stop assessments before systolic BP is dangerously increasing, thereby reducing the risk of potentially life-threatening complications in this cohort.

Wan D, Krassioukov A V. Life-threatening outcomes associated with autonomic dysreflexia: A clinical review. J Spinal Cord Med. 2014
Claydon VE, Elliott SL, Sheel AW, Krassioukov A. Cardiovascular responses to vibrostimulation for sperm retrieval in men with spinal cord injury. J Spinal Cord Med. 2006;29(3):207-216.
Solinsky R, Kirshblum SC, Burns SP. Exploring detailed characteristics of autonomic dysreflexia. J Spinal Cord Med. 2018;41(5):549-555.