1. Blood Pressure Measurement in SPRINT
Embargo lifts Mon., Nov. 13, 10:45 a.m. to 12 noon PT
Karen C. Johnson, MD, MPH, FAHA
Vice Chair, SPRINT Steering Committee
University of Tennessee Health Science Center, Department of Preventive Medicine
For the SPRINT Research Group
November 13, 2017
Good morning. Thank you for coming to the Blood Pressure Measurement in SPRINT presentation.

2. SPRINT Research Question
Examine effect of more intensive high blood pressure treatment
than is currently recommended
Randomized Controlled Trial
Target Systolic BP
Intensive Treatment
Goal SBP < 120 mm Hg
Standard Treatment
Goal SBP < 140 mm Hg
Primary Outcome Composite
MI, ACS, Stroke, CHF or
Death from CVD
SPRINT was a randomized, controlled, open-label trial that assigned participants to a target systolic BP goal of either <140 mm Hg (Standard) or <120 mm Hg (Intensive). The SPRINT trial was conducted from November 2010 until August 2015 in 102 clinical sites organized into five clinical center networks.
The primary hypothesis for SPRINT was that treatment to a SBP target of <120 mmHg compared to a SBP target of <140 mmHg would reduce the composite primary outcome of acute myocardial infarction (MI), acute coronary syndrome not resulting in MI (non-MI ACS), stroke, acute decompensated heart failure (HF), and death from CVD.
SPRINT design details available at:
ClinicalTrials.gov (NCT )
Ambrosius WT et al. Clin. Trials. 2014;11:

3. Blood Pressure Measurement in SPRINT
Similar to what has been used in virtually all recent HTN outcome trials.
Similar to what has been recommended for clinical practice by virtually all HTN guidelines
SPRINT Blood Pressure Measurement Procedures
SPRINT BP was the average of 3 BP measurements obtained using an automated measurement device (Omron 907XL) after a 5 minute rest period.
Appropriate cuff size was determined by measuring arm circumference.
Participant was seated with back supported and arm bared and supported at heart level.
Omron Device was set to delay 5 minutes and then take/average 3 BP measurements, during which time participants refrained from talking or texting.
Knowing how BP is measured is important to understanding BP control and guiding clinicians in appropriate management of hypertension.
The SPRINT trial use BP measurement methods similar to what have been used in recent HTN outcome trials and similar to what is recommended for clinical practice by virtually all HTN guidelines
99.7% of the BP measurements were taken in SPRINT with the study provided Omron device.
Cushman, et al. Hypertension. 2016;67:263-5

4. Background
The SPRINT clinical trial demonstrated that treatment to an intensive systolic BP goal (<120 mmHg) reduces risk of cardiovascular disease (CVD) and mortality compared with standard treatment (< 140 mm Hg)
Recent publications from investigators not involved with SPRINT have stated that the BP measurement technique used in SPRINT was unattended, making the BP not comparable to other studies.
SPRINT protocol does not address the issue of staff attendance, nor have SPRINT investigators yet published information on this topic.
As background for this presentation
The SPRINT trial was stopped early for benefit in 2015 because it demonstrated that ….
Recent publications from investigators not involved with SPRINT…
However, the SPRINT protocol does not address the issue of staff attendance…
Sprint Research Group. NEJM 2015;373:

5. Objectives
To present the results of a survey conducted immediately after SPRINT closeout visits were completed to inquire whether BP measurements were usually attended or unattended by staff.
To determine whether variation in blood pressure measurement technique was associated with differences in blood pressure obtained, medication utilization, CVD or mortality outcomes, or serious adverse events (SAE).
To assess whether BP measurements were attended or unattended at SPRINT clinics, we conducted a survey immediately after study closeout to inquire whether site staff were usually in the room with the participant (attended) or not in the room (unattended) during the rest period and/or during the BP measurement. We examined whether there were differences in measured clinic BP at baseline and during follow-up, and whether there were corresponding differences in SPRINT major outcome results or safety events, based on staff attendance during the measurement of BP.

6. BP Measurement Technique Categories
Always Alone (4082 participants at 38 sites)
Participant alone for 5 minutes of rest and the 3 BP readings
Never Alone (2247 participants at 25 sites)
Personnel in the room during the entire BP measurement period
Alone for Rest (1746 participants at 19 sites)
Participant alone only during the rest period
Alone for BP Measurement (570 participants at 6 sites)
Participant alone only during the BP readings
Excluded from analyses ( 716 participants at 14 sites)
Based on the survey …
Majority of coordinators were extremely confident or very confident in their responses to the survey

7. by Randomized Group and Blood Pressure Technique
Baseline Clinical Characteristics of SPRINT Participants Stratified
by Randomized Group and Blood Pressure Technique
Always Alone
(n =4082 )
Never Alone
(n =2247 )
Alone for Rest
(n =1746)
Alone for BP Measurement
(n=570 )
Characteristics
Intensive
Standard
No. Clinics 38 25 19 6
No. Randomized
2037
2045
1123
1124
875
871
283
287
Age %
35.8
37.5
50.3
49.1
34.6
34.0
46.6
45.3
65-74 %
33.1
32.1
26.5
25.8
32.7
33.0
32.9
36.2
≥75 %
31.1
30.4
23.2
25.1
20.4
18.5
Female Gender %
32.8
30.3
31.0
40.1
43.2
48.8
48.4
Black Race %
29.2
30.6
36.4
37.3
27.9
26.2
28.6
27.2
Hispanic Ethnicity %
3.1
2.2
17.2
16.2
3.0
64.7
65.2
Chronic kidney disease‡ %
31.2
24.0
29.0
20.1
23.3
Cardiovascular disease %
21.5
21.9
21.2
21.1
16.5
19.1
17.8
Framingham 10-yr CVD risk score mean
26.0
24.8
23.8
23.6
22.5
Not surprisingly, the 4 BP measurement technique categories differed by baseline characteristic.

8. by Randomized Group and Blood Pressure Technique
Baseline Clinical Characteristics of SPRINT Participants Stratified
by Randomized Group and Blood Pressure Technique
Always Alone
(n =4082 )
Never Alone
(n =2247 )
Alone for Rest
(n =1746)
Alone for BP Measurement (n=570 )
Characteristics
Intensive
Standard
No. Clinics 38 25 19 6
No. Randomized
2037
2045
1123
1124
875
871
283
287
Baseline BP
SBP mean, mmHg
139.6
139.9
139.2
138.8
140.0
142.5
141.1
DBP mean, mmHg
77.9
78.0
79.3
78.7
77.8
77.1
76.9
76.2
BMI mean, Kg/m2
30.0
29.9
30.2
29.3
29.2
29.6
29.4
Smoking Status
Current %
13.3
12.6
16.9
14.7
9.7
11.1
13.1
11.5
Past %
45.2
44.7
41.9
42.1
40.8
44.8
34.3
27.6
Statin use %
46.3
41.0
44.5
41.1
44.2
34.4
39.8
Aspirin use %
54.6
54.8
51.2
49.2
50.5
47.8
36.4
37.4

9. Systolic Blood Pressure by BP Measurement Technique
Alone for Rest
Always Alone
Average SBP Delta 13.1
Average SBP Delta 12.5
Never Alone
Alone for BP Measurement
The Intensive group in all four BP measurement categories achieved a systolic BP of 120 mm Hg whereas the Standard group achieved a systolic BP of 135.
The average systolic BP delta in all four BP measurement technique categories was very similar and differed by less than 2.3 mm Hg.
At the majority of follow-up visits, similar average number of medication use within randomized groups was noted among participants in all four BP measurement technique categories. On average the Intensive group was taking 1 additional medication than the Standard group in all four BP measurement technique categories.
Average SBP Delta 13.3
Average SBP Delta 14.7

10. Primary and all-cause mortality outcomes
stratified by treatment group and blood pressure technique
Intensive Arm
Standard Arm
Intensive vs. Standard Hazard Ratio
Outcome
BP Technique
Events
% Per year HR
95% CI
Interaction P-value
Primary
Always Alone
101
1.5
159
2.5
0.62
0.51,0.76
0.005
Never Alone 68
1.9
103
3.0
0.64
0.46,0.91
Alone for Rest 50
1.8 51
0.98
0.76,1.25
Alone for BP Measurement 20
2.1 15
1.39
0.78,2.49
Total Mortality 64
1.0 98
0.65
0.47,0.88
0.28 46
1.3 60
1.7
0.76
0.53,1.11 19
0.7 32
1.1
0.59
0.37,0.94 10 7
1.48
0.63,3.05
0.88
When all four measurement technique categories were compared, heterogeneity of treatment effect was detected for the primary outcome. In both the AA and NA categories the Intensive treatment group had a reduced risk of the primary outcome compared to the Standard treatment group; however, no significant difference between the treatment groups among participants in the smaller AR and the ABM categories was detected.
When the larger AA and NA groups were compared with one another, there was no evidence of heterogeneity of treatment effect by BP measurement technique for the primary outcome.
No heterogeneity of treatment effect was detected for total mortality by BP measurement technique
Primary Outcome Composite - MI, ACS, Stroke, CHF or Death from CVD

11. Serious Adverse Events (SAE) and other monitored adverse event
outcomes stratified by treatment group and blood pressure technique
Intensive Arm
Standard Arm
Intensive vs. Standard Hazard Ratio
Outcome
BP Technique
% Per year HR
95% CI
Interaction P-value
Any SAE
Always Alone
15.4
15.0
1.03
0.95,1.12
0.57
Never Alone
14.7
14.3
1.02
0.85,1.23
Alone for Rest
14.6
14.8
0.99
0.88,1.10
Alone for BP Measurement
8.9
7.7
1.16
0.96,1.41
Syncope
0.8
0.4
1.79
1.20,2.68
0.14
0.5
1.19
0.61,2.34
0.6
0.82
0.46,1.46
0.68
0.11,4.42
Hypotension
0.7
1.73
1.03,2.92
0.48
0.3
2.50
1.30,4.78
1.22
0.55,2.71
0.61
0.04,9.99
Injurious Fall
0.95
0.57,1.59
0.64
1.10
0.66,1.84
0.49,1.38
0.2
2.59
0.40,16.76
There was no difference in total SAEs between the Intensive and Standard groups in any of the four BP measurement technique categories and we could detect no significant heterogeneity of treatment effect.
I want to underscore the low rates of events for syncope, hypotension and injurious falls in both the Intensive and Standard groups in all 4 BP measurement technique categories. Further, no heterogeneity of treatment effect by BP measurement technique was detected.

12. Serious Adverse Events (SAE) and other monitored adverse event
outcomes stratified by treatment group and blood pressure technique
Intensive Arm
Standard Arm
Intensive vs. Standard Hazard Ratio
Outcome
BP Technique
% Per year HR
95% CI
Interaction P-value
Bradycardia
Always Alone
0.5
1.15
0.74,1.81 --
Never Alone
1.05
0.60,1.84
Alone for Rest
0.4
0.6
0.70
0.31,1.58
Alone for BP Measurement
Acute Kidney Injury
1.4
0.9
1.50
1.11,2.03
0.41
1.3
2.08
1.16,3.75
1.1
2.15
1.22,3.77
0.8
0.7
1.18
0.62,2.26
Electrolyte Abnormal 1
1.38
1.04,1.83
0.38
0.91
0.50,1.67
1.47
0.80,2.72
0.2
3.10
0.78,12.38
Low rates of bradycardia, acute kidney injury, and electrolyte abnormality were seen in both the Intensive and Standard arms and no heterogeneity of treatment effect was observed by BP measurement technique category.

13. Limitations Post-hoc survey to assess BP measurement technique
BP measurement technique classifications based on staff recall – potential for misclassification
Unattended and attended BP measurements were not made in the same individual
Survey was not prespecified prior to the beginning of the trial
Limitations of these data include….

14. Summary and Conclusions
Similar BP levels and CVD risk reduction were observed in the Intensive group whether the BP measurements technique used was primarily attended (NA) or primarily unattended (AA).
To fully realize the benefits and minimize risks associated with following the SPRINT Intensive treatment algorithm, use of a validated automated BP device, staff training to allow for a quiet rest period, proper positioning, use of proper cuff size, and averaging multiple measurements, may be more important than whether the BP measurement is attended or unattended.
To arrive at firmer conclusions, additional research with better methods, is needed to determine whether attendance or other factors during the BP measurement affect level of BP reading.
Blood pressure in SPRINT was measured in a standardized method in both attended and unattended conditions. Similar SBP differences between the Intensive and Standard groups were seen in all four BP measurement technique categories regardless of staff attendance and these differences were well maintained over the course of the trial.
CVD and total mortality risk were reduced similarly in participants treated to the Intensive compared with Standard goals in clinics using either the NA and AA BP measurement techniques, with no evident increased risk of SAEs related to whether BP measurements were attended or unattended.
To arrive at firmer conclusions, additional research with better methods, is needed to determine whether attendance or other factors during the BP measurement affect level of BP reading.

15. Acknowledgements 9,361 volunteers who agreed to participate in SPRINT
Investigators and staff, including Steering Committee, other principals at the 5 Clinical Center Networks, 102 participating Clinical Centers, Coordinating Center, Central Laboratory, ECG Reading Center, MRI Reading Center, and Drug Distribution Center
National Institutes of Health
National Heart, Lung, and Blood Institute (NHLBI)
National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK)
National Institute on Aging (NIA)
National Institute of Neurological Disorders and Stroke (NINDS)
Takeda and Arbor Pharmaceuticals (donated 5% of medication used)
I would like to thank the 9361 volunteers who participated in SPRINT along with the investigators and staff.
I would like to thank the NIH for sponsoring the trial including…
Thank you for you attention.