Heated debate about interpretation of SPRINT-data at ESC congress
Sources
News - Sep. 7, 2016The SPRINT study, which was terminated early due to favourable results (published in NEJM in November 2015 [1]), was the topic of an interesting pro-con debate during the ESC congress, in Rome on August 28.
In light of long known discussions on what is the best blood pressure (BP) target for CV prevention, the Systolic Blood Pressure Intervention Trial (SPRINT) (n=9361) was an interesting study. Older (>50 years) non-diabetic patients with a systolic BP (SBP) between 130 and 180 mmHg and at least one CVD risk factor were randomised to a target of 120 mmHg or 140 mmHg (standard in this age group; control group)(non-blinded study design, median follow-up: 3.26 years).
Targets were well met and in the group receiving intensive blood pressure lowering treatment the primary composite endpoint of first occurrence of myocardial infarction (MI), non-MI ACS, stroke, decompensated heart failure or CVD mortality, was seen 25% less often than in the control group. The lower risk was mainly driven by a reduction in heart failure and CVD mortality of about 40% each.
The debate concerned the question whether the SPRINT results should influence recommended targets in new guidelines. While professor William C Cushman (Memphis, TN, USA) defended the results and argued for lowering of targets based on the SPRINT results, professor Sverre E Kjeldsen (Oslo, Norway) expressed some critical concerns he had with the study design and outcomes.
Firstly, he considered the method of BP measurement. In SPRINT, an ‘unattended’ automated method was applied. A patient was left alone in a room and in absence of a care professional an automated oscillometric sphyghmomanometer measured BP three times (AOBP), five minutes after the health professional has left the room, with intervals of 1 minute. While AOBP methods have been applied in other studies, SPRINT is the only trial in which the patient was left alone during the measurement (unattended). In the origininal NEJM publication, this aspect was not mentioned.
It is known that AOBP yields a lower BP than standard measurement by a clinician or nurse, because the white coat effect is minimised and human interaction/interpretation is limited. It has also been shown that the presence of a clinician, and to a lesser extent a nurse, in the room can directly affect BP [2]. Kjeldsen therefore argued that an SBP of 120 mmHg measured in SPRINT can be compared with 140 mmHg in other studies.
Cushman had mentioned about this that it is highly variable from patient to patient how the method of measuring affects the measurement.
In a perspective in Circulation [3], George L Bakris wrote that current standard methods seem to overestimate blood pressure. Improving the accuracy of office BP measurement like with AOBP is a goal. If, however, the patient gets no resting period and the automatic repeated measurements cannot be performed unattendedly, a 5-10 mmHg SBP higher target than in a trial like SPRINT should be applied.
During the debate, Cushman further noted that the most important finding, namely the difference in incidence of heart failure, may have been the result of the uptitration and downtitration of diuretics in the intensively treated and the control group, respectively. In the control group diuretics were frequently discontinued, to keep BP around the target of 140 mmHg. It is possible that this has masked or unmasked the endpoint in the respective groups, in these patients with a high risk of heart failure.
Cushman had shown that there was no overall difference between treatment groups in the number of serious adverse events (SAEs). SAEs associated with hypotension, syncope, electrolyte disturbances and hospital discharge records of acute renal injury or acute renal failure occurred more often in the intensively treated group. Nevertheless, Cushman concluded that the benefits of intensive blood pressure-lowering outweighed the risk of harm.
Kjeldsen emphasised, on the other hand, that he thought the number of SAEs was rather high in the intensive treatment arm; there were many unfavourable and potentially harmful SAEs. He referred to a piece in Ann Intern Med earlier this year in which it was calculated that the risk of harm (22/1000 patients) was larger than the risk of benefit of treatment (16/1000)[4].
Kjeldsen was therefore convinced that SPRINT results should not be used as a reason to lower targets in the guidelines. During a voting session in the audience of the debate it appeared that the majority were mostly convinced by the arguments brought up by Kjeldsen.
In a concluding remark of the discussion, professor Bryan Williams concluded that “there is little doubt that more BP lowering is better for patients at high risk.” The debate is about how low one should go. His own view is that “guidelines should always remain conservative, in the sense that you cannot advocate levels of treatment that are potentially harmful to some, even if it benefits others.”
Then on an individual basis, a clinician may estimate, based on the evidence, that going further may be beneficial for a given patient. He also brought up that maybe clinicians need to consider working with target ranges, especially when recommended targets are lower than what is conventionally applied.
Williams concluded with the sobering remark that, however relevant the discussion of how low to go in some high-risk patients, a large part of the patients does not reach current recommendations of <140 mmHg. So along with continuing the debate, efforts should be directed at more effectively treating everybody.
ESC Debat 28 augustus 2016, online available.
In light of long known discussions on what is the best blood pressure (BP) target for CV prevention, the Systolic Blood Pressure Intervention Trial (SPRINT) (n=9361) was an interesting study. Older (>50 years) non-diabetic patients with a systolic BP (SBP) between 130 and 180 mmHg and at least one CVD risk factor were randomised to a target of 120 mmHg or 140 mmHg (standard in this age group; control group)(non-blinded study design, median follow-up: 3.26 years).
Targets were well met and in the group receiving intensive blood pressure lowering treatment the primary composite endpoint of first occurrence of myocardial infarction (MI), non-MI ACS, stroke, decompensated heart failure or CVD mortality, was seen 25% less often than in the control group. The lower risk was mainly driven by a reduction in heart failure and CVD mortality of about 40% each.
The debate concerned the question whether the SPRINT results should influence recommended targets in new guidelines. While professor William C Cushman (Memphis, TN, USA) defended the results and argued for lowering of targets based on the SPRINT results, professor Sverre E Kjeldsen (Oslo, Norway) expressed some critical concerns he had with the study design and outcomes.
Firstly, he considered the method of BP measurement. In SPRINT, an ‘unattended’ automated method was applied. A patient was left alone in a room and in absence of a care professional an automated oscillometric sphyghmomanometer measured BP three times (AOBP), five minutes after the health professional has left the room, with intervals of 1 minute. While AOBP methods have been applied in other studies, SPRINT is the only trial in which the patient was left alone during the measurement (unattended). In the origininal NEJM publication, this aspect was not mentioned.
It is known that AOBP yields a lower BP than standard measurement by a clinician or nurse, because the white coat effect is minimised and human interaction/interpretation is limited. It has also been shown that the presence of a clinician, and to a lesser extent a nurse, in the room can directly affect BP [2]. Kjeldsen therefore argued that an SBP of 120 mmHg measured in SPRINT can be compared with 140 mmHg in other studies.
Cushman had mentioned about this that it is highly variable from patient to patient how the method of measuring affects the measurement.
In a perspective in Circulation [3], George L Bakris wrote that current standard methods seem to overestimate blood pressure. Improving the accuracy of office BP measurement like with AOBP is a goal. If, however, the patient gets no resting period and the automatic repeated measurements cannot be performed unattendedly, a 5-10 mmHg SBP higher target than in a trial like SPRINT should be applied.
During the debate, Cushman further noted that the most important finding, namely the difference in incidence of heart failure, may have been the result of the uptitration and downtitration of diuretics in the intensively treated and the control group, respectively. In the control group diuretics were frequently discontinued, to keep BP around the target of 140 mmHg. It is possible that this has masked or unmasked the endpoint in the respective groups, in these patients with a high risk of heart failure.
Cushman had shown that there was no overall difference between treatment groups in the number of serious adverse events (SAEs). SAEs associated with hypotension, syncope, electrolyte disturbances and hospital discharge records of acute renal injury or acute renal failure occurred more often in the intensively treated group. Nevertheless, Cushman concluded that the benefits of intensive blood pressure-lowering outweighed the risk of harm.
Kjeldsen emphasised, on the other hand, that he thought the number of SAEs was rather high in the intensive treatment arm; there were many unfavourable and potentially harmful SAEs. He referred to a piece in Ann Intern Med earlier this year in which it was calculated that the risk of harm (22/1000 patients) was larger than the risk of benefit of treatment (16/1000)[4].
Kjeldsen was therefore convinced that SPRINT results should not be used as a reason to lower targets in the guidelines. During a voting session in the audience of the debate it appeared that the majority were mostly convinced by the arguments brought up by Kjeldsen.
In a concluding remark of the discussion, professor Bryan Williams concluded that “there is little doubt that more BP lowering is better for patients at high risk.” The debate is about how low one should go. His own view is that “guidelines should always remain conservative, in the sense that you cannot advocate levels of treatment that are potentially harmful to some, even if it benefits others.”
Then on an individual basis, a clinician may estimate, based on the evidence, that going further may be beneficial for a given patient. He also brought up that maybe clinicians need to consider working with target ranges, especially when recommended targets are lower than what is conventionally applied.
Williams concluded with the sobering remark that, however relevant the discussion of how low to go in some high-risk patients, a large part of the patients does not reach current recommendations of <140 mmHg. So along with continuing the debate, efforts should be directed at more effectively treating everybody.
ESC Debat 28 augustus 2016, online available.
1. The SPRINT research group. A Randomized Trial of Intensive versus Standard Blood-Pressure Control. N Engl J Med 2015; 373:2103-2116
2. Filipovský J et al., Automated compared to manual office blood pressure and to home blood pressure in hypertensive patients. Blood Press. 2016 Aug;25(4):228-34.
3. Bakris GL. The Implications of Blood Pressure Measurement Methods on Treatment Targets for Blood Pressure. Circulation. Published Ahead of Print: August 30, 2016 http://dx.doi.org/10.1161/CIRCULATIONAHA.116.022536
4. Ortiz E and James PA. Let's Not SPRINT to Judgment About New Blood Pressure Goals. Ann Intern Med. 2016;164(10):692-693.
Facebook Comments