Mean Arterial Pressure Targets and Patient-Important Outcomes in Critically Ill Adults: A Systematic Review and Meta-Analysis of Randomized Trials* : Critical Care Medicine

Journal Logo

Feature Articles

Mean Arterial Pressure Targets and Patient-Important Outcomes in Critically Ill Adults: A Systematic Review and Meta-Analysis of Randomized Trials*

Carayannopoulos, Kallirroi Laiya MD, FRCPC1; Pidutti, Andrew BSc2; Upadhyaya, Yashita MD1; Alshamsi, Fayez MD, FRCPC3; Basmaji, John MD, FRCPC4; Granholm, Anders MD5; Alhazzani, Waleed MD, MSc, FRCPC1,6; Lewis, Kimberley MD, MSc, FRCPC1,6

Author Information
Critical Care Medicine 51(2):p 241-253, February 2023. | DOI: 10.1097/CCM.0000000000005726

Abstract

KEY POINTS

  • Question: In critically ill adults with shock, does targeting a higher blood pressure using vasoactive medications improve patient outcomes?
  • Findings: In this systematic review and meta-analysis, we found targeting a higher MAP (75–85 mm Hg) compared with lower MAP of 65 mm Hg likely resulted in no difference in mortality (relative risk, 1.06; 95% CI, 0.98–1.15; I2 = 0%; p = 0.12; moderate certainty).
  • Meaning: There is no difference in mortality when a higher MAP is targeted in critically ill adult patients with shock.

Hypotension is a common and potentially catastrophic complication of critical illness. Persistent and severe systemic hypotension ultimately leads to tissue hypoperfusion and end-organ damage, perpetuating critical illness and ischemia (1).

Achieving a higher mean arterial pressure (MAP) theoretically improves perfusion and therefore patient outcomes, and fluids and vasoactive drugs are the mainstay therapies to accomplish this. The 2021 Surviving Sepsis Campaign Guidelines (2) strongly recommend targeting a MAP of 65 mm Hg through the use of fluids and vasopressors in patients with septic shock. Both interventions, however, do not come without risk. Overzealous fluid administration has been associated with prolonged mechanical ventilation and ICU length of stay, increased acute kidney injury, and mortality (3). Vasopressor use must be balanced against the risk of arrhythmia and ischemic events (4).

Several systematic reviews (5,6) examining optimal blood pressure targets in critically ill patients with varying etiologies of shock have been published without finding certain evidence for meaningful difference in mortality with higher MAP targets; however, they included a limited number of studies and were conducted prior to the publication of the most recent randomized controlled trials (RCTs) on this topic (7–9). Considering this, we conducted a systematic review and meta-analysis to address the potential benefits and harms of targeting higher versus lower MAP.

MATERIALS AND METHODS

This study was prospectively registered PROSPERO: CRD42020219971.

Study Selection

We included parallel-group RCTs that included adult patients (≥18 yr old) with a diagnosis of shock requiring vasoactive medications. The intervention group was required to receive vasoactive medications at any dose, frequency, initiation time, or duration of treatment to target a higher MAP as established by study authors, whereas the control group received vasoactive medications with the same criteria to target a lower MAP. Eligible studies reported on at least one of: mortality at longest follow-up (primary outcome) and 28 days, duration of vasopressor therapy, volume of fluid administered, need for renal replacement therapy, ICU length of stay, hospital length of stay, health-related quality of life at longest follow-up, or adverse events. Post hoc, we elected to include two studies with a reversal of the study populations (i.e., higher MAP target as control and lower MAP target as intervention) by using the data in reverse (i.e., lower MAP as control and higher MAP as intervention) as we felt it was still reflective of the goals of the meta-analysis and would be a significant loss to not include a large study population in the data analysis.

Search Strategy

A medical librarian performed an electronic search of MEDLINE, EMBASE, and the Cochrane Library from inception to May 12, 2021, without language restriction. We searched ongoing trials in ClincialTrials.gov to May 12, 2021 (Supplement Tables 1–5, https://links.lww.com/CCM/H247). Finally, we screened the reference list of relevant review articles for additional studies.

Selection of Trials

Five reviewers screened titles and abstracts independently and in duplicate to identify potentially eligible studies, then conducted full-text evaluation for final eligibility. Disagreements between reviewers were resolved through consultation with a third reviewer.

Data Abstraction

In triplicate, reviewers independently extracted data using a prepiloted abstraction form. Disagreements were resolved by discussion and consensus.

Risk of Bias

Two reviewers independently assessed trials for risk of bias using the Cochrane risk of bias tool 2.0 (10). For each included trial, we judged individual outcomes as low, some concern, or high risk of bias in the domains of the randomization process, deviation from intended intervention, missing outcome data, measurement of the outcome, and selection of reported results. Disagreements were resolved by discussion to reach consensus.

Statistical Analysis

Statistical analyses were conducted using the RevMan software Version 5.3 (11) and the trial sequential analysis (TSA) Software (Version 0.9.5.10 beta, Copenhagen Trial Unit, ctu.dk/tsa). We used the DerSimonian and Laird random-effects model to pool the weighted effect of estimates across all studies (12). The inverse variance method was used to estimate study weights. We calculated pooled relative risks (RRs) for dichotomous outcomes and mean differences (MDs) for continuous outcomes, with corresponding 95% CIs. We planned to inspect funnel plots to assess for publication bias if more than ten trials existed (13) for a given outcome.

Trial Sequential Analysis

We used TSA to determine if the required information size to reach firm conclusions was met in order to reduce the risk of spurious findings (13–15). TSA estimates the information size required to firmly accept or refute a certain prespecified effect size of interest. TSA calculates cumulative z scores and boundaries for superiority, inferiority, or futility; when any of the boundaries is crossed before the required information size has been reached, evidence may be considered conclusive for the targeted effect size (16). If the cumulative z-curve does not cross any of the threshold boundaries, we conclude that the quantity of evidence is insufficient to firmly accept or refute the effect size of interest (14,15). TSA was conducted using an alpha of 0.05, power of 0.90 (beta 0.10), and diversity (D2) (16), as suggested by the included trials, and unweighted control event proportions as per the included trials for binary outcomes and variances as estimated in the included trials for continuous outcomes. Effect sizes of interest were determined post hoc (Supplement Table 6, https://links.lww.com/CCM/H247). Full TSA results are presented in Supplement Table 7 and Supplement Figures 1–7 (https://links.lww.com/CCM/H247).

Heterogeneity and Subgroup Analysis

Statistical heterogeneity was assessed using the chi-square test, I2 statistics, and visual inspection of the forest plots (17). Heterogeneity between studies was explored by performing predefined subgroup analyses. These subgroups included: 1) history of chronic hypertension versus no documented history, 2) cause of shock (distributive vs hemorrhagic vs cardiogenic), 3) type of vasopressor (epinephrine vs norepinephrine vs phenylephrine vs other), and 4) age greater than 65 versus less than or equal to 65 years. An additional subgroup analysis was conducted post hoc based on control MAP targets (standard of care [>65 mm Hg] or lower [i.e., ≤65 mm Hg]). Prespecified hypothesized directions of effect for all subgroups are available in Supplement Table 8 (https://links.lww.com/CCM/H247) and protocol (18).

A sensitivity analysis was planned to explore the impact on the pooled results by restricting analyses to studies adjudicated as low risk of bias only. Sensitivity analyses excluding the single study with control MAP target less than 60 mm Hg as well as the two studies with control target less than 65 mm Hg were performed post hoc.

Assessing the Certainty of Evidence

Two reviewers independently and in duplicate used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to assess the certainty of evidence for each outcome (19). We used the GRADEpro software (GRADEpro GDT: GRADEpro Guideline Development Tool [Software], McMaster University, 2020) (20) to create the evidence profile (21). Justification for the downgrading of certainty of evidence for each outcome is available in Supplement Table 19 (https://links.lww.com/CCM/H247).

RESULTS

Screening

Our electronic search identified 14,702 citations (Fig. 1), of which 12,212 remained after removal of duplicates. Title and abstract screening resulted in 91 studies undergoing full-text review, of which 85 were then excluded (Supplement Table 9, https://links.lww.com/CCM/H247). The remaining six studies were deemed eligible and were included in the final analysis.

F1
Figure 1.:
Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart.

Characteristics of Included Studies

Overall, six studies (n = 3,690) (7–9,22–24) met eligibility criteria (Table 1). The mean age was 69.9 ± 15.1 (sd) years, and 36% of enrolled patients were female. Forty-four percent of patients had a documented preexisting diagnosis of hypertension. The mean Acute Physiologic Assessment and Chronic Health Evaluation II score in the studies (7,24) that reported was 20.9 ± 6.4, and the Sequential Organ Failure Assessment score in the studies (8,22) that reported was 10.5 ± 3.1 (calculated using pooled variance). Three studies (7,22,24) (n = 3,357) exclusively enrolled patients whose cause of shock was distributive, whereas two studies enrolled patients with cardiogenic shock (n = 169). The one remaining study (23) enrolled patients with hemorrhagic shock only (n = 164).

TABLE 1. - Characteristics of Included Studies
References Population Intervention Control Primary Outcome Follow-Up Funding
Demographics Inclusion Criteria
Lamontagne et al (7), n = 2,463 Median age 75.0 (± 7.7) Adults (≥ 65 yr) admitted to a participating ICU, randomized within 6 hr of commencing a vasopressor infusion for vasodilatory hypotension, with adequate fluid resuscitation completed or ongoing and vasopressors expected to continue for 6 hr or more Vasopressors at the discretion of treating clinicians, personalized approach (e.g., in function of patient characteristics and markers of perfusion) MAP target permissive hypotension (60–65 mm Hg). Choice of vasopressor, as well as all other interventions at the discretion of treating clinicians All-cause mortality at 90 d 90 d NIHR Health Technology Assessment Programme. The Intensive Care National Audit & Research Centre NIHR Research Professorship award and NIHR Comprehensive Biomedical Research Centre. FRQS career award
n female = 1,067
APACHE II = 20.6 (± 6.3)
n chronic HTN = 1,187
Lamontagne et al (24), n = 118 Median age 64.5 (± 13.0) Adults (≥ 16 yr) receiving vasopressors in ICU for presumed vasodilatory shock, if treating physician judged that they were adequately fluid resuscitated and that ongoing vasopressor therapy was expected for at least 6 hr MAP targeted to 75–80 mm Hg. When ICU physicians judged patient no longer in need of vasopressor therapy, the infusion was stopped even if MAP was no longer in range MAP targeted to 60–65 mm Hg. When ICU physicians judged patient no longer in need of vasopressor therapy, the infusion was stopped even if MAP was no longer in range Hospital mortality 6 mo The Canadian Institutes for Health Research and the FRQS
n female = 54
APACHE II = 24.5 (± 7.1)
n chronic HTN = 53
Grand et al (9), n = 49 Median age 60.9 (± 11.7) Adults (≥ 18 yr), comatose (Glasgow Coma Score ≤ 8) resuscitated out of hospital cardiac arrest patients of presumed cardiac cause with sustained ROSC for > 20 min, irrespective of the initial rhythm Blindly assigned modules with a 10% offset, resulting in 72 mm Hg MAP target. Norepinephrine utilized MAP target of 65 mm Hg, norepinephrine utilized. Plasma concentration of soluble thrombomodulin after 48 hr 180 d None
n female = 6
APACHE II = NR
n chronic HTN = 20
Asfar et al (22), n = 776 Median age 65.0 (± 14.0) Adults (≥ 18 yr) with septic shock refractory to fluid resuscitation, requiring vasopressors (norepinephrine or epinephrine) at a minimum infusion rate of 0.1 μg/kg/min, and evaluated within 6 hr after the initiation of vasopressors MAP targeted to 80–85 mm Hg with norepinephrine/epinephrine MAP targeted to 65–70 mm Hg with norepinephrine/epinephrine Mortality at 28 d 90 d French Ministry of Health
n female = 259
SOFA = 10.8 (± 3.1)
n chronic HTN = 340
Carrick et al (23), n = 168 Median age 30.0 (± 28.5) All penetrating trauma patients, 14–45 yr old, with documented systolic blood pressure of 90 mm Hg or lower who were brought emergently to the operating room from the trauma bay for a laparotomy or thoracotomy to control bleeding MAP target of 65 mm Hg. MAPs represent the minimum blood pressures at which further interventions (fluids, transfusions, or vasopressors) were administered MAP target of 50 mm Hg. MAPs represent the minimum blood pressures at which further interventions (fluids, transfusions, or vasopressors) were administered. Blood pressure was never intentionally lowered All-cause mortality 30 d None
n female = 16
APACHE II = NR
n chronic HTN = NR
Ameloot et al (8), n = 120 Median age 62.5 (± 10.5) Adults (≥ 18 yr), resuscitated from out-of-hospital cardiac arrest of a presumed cardiac cause and unconscious at hospital admission after sustained ROSC. This study only included patients with shock after AMI The 36-hr intervention period started at ICU admission. Norepinephrine infusion and fluid boluses as needed to reach the assigned MAP target of 80–100 mm Hg The 36-hr intervention period started at ICU admission. MAP target of 65 mm Hg. Did not lower blood pressure by any means other than sedation and pain medication Myocardial injury 180 d
n female = 16
SOFA = 9.0 (± 2.6)
n chronic HTN = 54
APACHE = Acute Physiology and Chronic Health Evaluation, FRQS = Fonds de Recherche du Québec–Santé, HTN = hypertension, NIHR = National Institute for Health Research, MAP = mean arterial pressure, NR = not reported, ROSC = return of spontaneous circulation, SOFA = Sequential Organ Failure Assessment.

The higher MAP target ranged from 65 to 80 mm Hg with studies targeting a MAP greater than 65 mm Hg (23), greater than 72 mm Hg (9), greater than 75 mm Hg (24), and greater than 80 mm Hg (8,22), and the final study’s higher MAP target as per the clinician’s discretion using perfusion markers and patient characteristics, however, was over 65 mm Hg (7). For the lower MAP target, most studies targeted an MAP of 65 mm Hg (four (7–9,24) targeted 65 mm Hg or less and one (22) set 65–70 mm Hg as its lower target). One study targeted a lower MAP of 50 mm Hg (23). All studies allowed both vasopressor and inotropic agents. Three studies (7,23,24) did not provide any direction with respect to choice of vasopressor agent and left it to the clinician’s discretion. One study (8) allowed the use of norepinephrine and dobutamine through a targeted protocol, whereas another (9) allowed only norepinephrine and dopamine. Finally, one study (22) achieved target MAP through the combination of norepinephrine and epinephrine.

Risk of Bias

All six studies (7–9,22–24) were deemed to have a low risk of bias in all domains for all outcomes (see Supplement Tables 10–18, https://links.lww.com/CCM/H247, for full justification). Although only one study (9) was double blinded, all were deemed to be low risk as there were no deviations that arose as a result of the trial context, and there were balanced nonprotocol interventions. Outcome data were available for all participants in all studies for our outcomes of interest, and the measurement of the outcomes was not believed to be influenced by assessors’ knowledge of the assigned intervention. Finally, all six studies reported results and conducted their analyses in accordance with prespecified plans.

Mortality

Six studies (7–9,22–24) enrolling a total of 3,690 patients reported mortality at longest follow-up. Targeting a higher MAP compared with a lower MAP results in no difference in the risk of all-cause mortality at longest follow-up (RR, 1.06; 95% CI, 0.98–1.15; moderate certainty) (Fig. 2). A subgroup analysis comparing a control group target MAP less than or equal to 65 mm Hg versus greater than 65 mm Hg found no significant interaction (Supplement Fig. 8, https://links.lww.com/CCM/H247). A further subgroup analysis did not identify any interactions based on type of shock (Supplement Fig. 9, https://links.lww.com/CCM/H247). Subgroup analysis based on age and type of vasopressor could not be performed, nor could analysis by chronic hypertension status as data were available for this population in only one study.

F2
Figure 2.:
Forest plot showing mortality at longest follow-up (grouped by duration of follow-up). df = degrees of freedom, IV = inverse variance, MAP = mean arterial pressure.

Three studies (7,22,23) enrolling a total of 3,403 patients reported mortality at 28–30 days. Targeting a higher MAP resulted in no difference in the risk of all-cause mortality at days 28–30 in patients with shock (RR, 1.07; 95% CI, 0.98–1.18; moderate certainty) (Supplement Fig. 9, https://links.lww.com/CCM/H247). No statistically significant subgroup interaction was found with a lower MAP target of 65 mm Hg (Supplement Fig. 10, https://links.lww.com/CCM/H247).

Renal Replacement Therapy

Pooled analysis of three studies (7,9,22) (n = 3,285 participants) demonstrated no effect on risk of undergoing renal replacement therapy when targeting a higher MAP (RR, 0.96; 95% CI, 0.83–1.11; moderate certainty) (Fig. 3A). The subgroup analysis of studies with control MAP less than or equal to 65 mm Hg versus greater than 65 mm Hg did not find significant interaction (Supplement Fig. 11, https://links.lww.com/CCM/H247). Subgroup analysis of patients with a history of chronic hypertension captured two studies (7,22) and demonstrated a decreased risk of undergoing renal replacement therapy when a higher MAP was targeted (RR, 0.83; 95% CI, 0.71–0.98) (Supplement Fig. 12, https://links.lww.com/CCM/H247). Subgroup interaction could not be assessed due to lack of disaggregated data for patients without a history of chronic hypertension. Finally, there were no significant subgroup interactions based on type of shock (Supplement Fig. 13, https://links.lww.com/CCM/H247). Subgroup analyses based on age and type of vasopressor could not be completed due to insufficient data.

F3
Figure 3.:
A, Forest plot showing risk of undergoing renal replacement therapy. B, ICU length of stay (d). C, Hospital length of stay (d). D, Duration of vasopressor therapy (hr). df = degrees of freedom, IV = inverse variance, MAP = mean arterial pressure.

ICU Length of Stay

Three studies (7,9,22) reported ICU length of stay. Where available, data were used for all participants, but otherwise, the length of stay for survivors was collected, for a total of 2,540 participants, demonstrating no important effect on ICU length of stay with a higher MAP target (MD, 0.17 d; 95% CI, –0.30 to 0.63; high certainty) (Fig. 3B). Subgroup analyses by control group MAP target and type of shock did not demonstrate a significant subgroup interaction (Supplement Figs. 14 and 15, https://links.lww.com/CCM/H247). No further subgroup analysis by age, type of vasopressor, or history of chronic hypertension could be completed due to limited data.

Hospital Length of Stay

A pooled analysis of two studies (7,9) (n = 1,502) demonstrated no effect on hospital length of stay (MD, –0.81 d; 95% CI, –3.96 to 2.34; moderate certainty) (Fig. 3C). Subgroup analysis by shock type did not identify any statistically significant subgroup interaction (Supplement Fig. 16, https://links.lww.com/CCM/H247). There were insufficient studies and data for subgroup analyses by control MAP target, age, history of chronic hypertension, or type of vasopressor used.

Duration of Vasopressor Therapy

Four studies (7,9,22,24) enrolling a total of 3,480 patients reported duration of vasopressor therapy and demonstrated targeting a higher MAP increases duration of vasopressor therapy (MD, 17.12 hr; 95% CI, 0.60–33.63; low certainty) (Fig. 3D). A subgroup analysis of studies by control MAP target and type of shock did not identify any significant subgroup interactions (Supplement Figs. 17 and 18, https://links.lww.com/CCM/H247). Subgroup analysis based on age, type of vasopressor, or history of chronic hypertension could not be completed due to insufficient data.

Total Fluid Administered

Three studies (22–24) were included in the assessment of volume of fluid administered (n = 1045). There was no difference in total fluid administered regardless of higher or lower MAP target (MD, –0.01 L; 95% CI, –0.42 to 0.40 L; high certainty) (Supplement Fig. 19, https://links.lww.com/CCM/H247). Subgroup analysis by shock type identified no statistically significant interactions (Supplement Fig. 20, https://links.lww.com/CCM/H247). No subgroup analysis by MAP target, history of chronic hypertension, age, or type of vasopressor used could be completed due to lack of studies and adequate data.

Digital Ischemia or Necrosis

Three studies (7,22,24) were included, enrolling a total of 3,476 patients and demonstrating no effect on the risk of digital ischemia or necrosis (RR, 0.89; 95% CI, 0.43–1.84; moderate certainty) (Fig. 4A). Subgroup analysis based on control MAP did not identify any significant subgroup effects (Supplement Fig. 21, https://links.lww.com/CCM/H247). There were insufficient data to perform the subgroup analyses by type of shock, age, type of vasopressor, or history of chronic hypertension.

F4
Figure 4.:
A, Forest plot showing risk of digital ischemia/necrosis. B, Risk of mesenteric ischemia. C, Incidence of cardiac arrhythmias. df = degrees of freedom, IV = inverse variance, MAP = mean arterial pressure.

Mesenteric Ischemia

Pooled analysis of three studies (7,22,24) (n = 3,476) demonstrated no difference in risk of mesenteric ischemia (RR, 1.09; 95% CI, 0.60–1.98; moderate certainty) (Fig. 4B). No statistically significant subgroup interactions were found based on a control MAP target less than or equal to 65 mm Hg compared with over 65 mm Hg (Supplement Fig. 22, https://links.lww.com/CCM/H247). There were insufficient studies to complete further analyses based on shock type, age, type of vasopressor, or history of chronic hypertension.

Cardiac Arrhythmias

Four studies (7,8,22,24) captured incidence of cardiac arrhythmias for a total of 3,596 participants. There was no difference in the risk of arrhythmic events (RR, 1.34; 95% CI, 0.83–2.19; low certainty) (Fig. 4C). Subgroup analysis based on shock type did not identify any significant subgroup interaction (Supplement Fig. 23, https://links.lww.com/CCM/H247). Further subgroup analysis based on control MAP target, age, history of chronic hypertension, or type of vasopressor could not be completed due to insufficient data.

Sensitivity Analysis

Sensitivity analysis excluding unclear or high risk of bias studies was not conducted as all included studies were deemed to have low risk of bias. Post hoc sensitivity analyses excluding one study (23) that used a lower MAP target of 50 mm Hg did not alter the results (Supplement Table 20, https://links.lww.com/CCM/H247) nor did the exclusion of two studies (7,23) with a lower MAP target below 65 mm Hg (Supplement Table 21, https://links.lww.com/CCM/H247).

DISCUSSION

In this systematic review and meta-analysis, we included six studies (7–9,22–24) (n = 3,690) comparing higher versus lower MAP targets in adult patients with shock. Overall, we found no difference in risk of mortality (moderate certainty), renal replacement therapy (moderate certainty), and digital or mesenteric ischemia (moderate certainty), or cardiac arrhythmias (low certainty). We found no reductions in the volume of fluid administered (high certainty), ICU length of stay (high certainty), or hospital length of stay (moderate certainty). Targeting a higher MAP likely increases the duration of vasopressor use (moderate certainty). A subgroup analysis of patients with chronic hypertension suggested decreased risk of undergoing renal replacement therapy when patients are treated with a higher MAP target.

Our findings are largely consistent with a previous systematic review (6), which found no effect on mortality, renal replacement therapy, digital ischemia, or mesenteric ischemia with higher MAP target. Our review is also consistent with a previous individual patient data meta-analysis (25) in patients with distributive shock that identified no difference in 28-day mortality with differing MAP targets. Although this analysis did identify a potential increased mortality risk in patients targeted to a higher MAP with vasopressor use over 6 hours, we were unable to examine outcomes stratified to duration or initiation time of vasopressors. Where our review also differs is that we did not identify any certain difference in risk of cardiac arrhythmias (5). This difference may be representative of a type 1 error as the previous review analyzed two small studies (22,24) and did not incorporate the largest study (7) to date, which was captured in our review. This may also be the result of our grouping of ventricular and supraventricular arrhythmias in the analysis rather than distinguishing the two, as the prior review (6) found no difference in ventricular arrhythmias but an increase in supraventricular arrhythmias with a higher MAP target.

The potential decreased risk of undergoing renal replacement therapy with a higher MAP target in patients with chronic hypertension—while only hypothesis-generating—seems biologically plausible. It is consistent with subgroup analysis of previous studies (22), and our knowledge of hypotension and its relationship to acute kidney injury. Although previous studies (26,27) assessing MAP and its relationship to urine output and serum creatinine have not demonstrated correlation, a prospective study (28) of patients in septic shock found that hypotension was associated with increased incidence of acute kidney injury. The potential decrease in need for renal replacement therapy with a higher MAP target in this population must be assessed in the context of a more pronounced increase in mortality observed in this subgroup in the largest study to date (7) on permissive hypotension as the potential increase in mortality may compromise the observed decrease in need for renal replacement therapy.

There are multiple strengths to this review. We undertook a methodologically rigorous process, adhered to an a priori registered protocol, performed an extensive search of the literature, and duplicated all aspects of this review. Finally, the conduction of TSA added strength and validated the need for further large-scale studies in this topic. This review is the first to include the most recent RCT on this topic (7).

Our review does have limitations. The actual average MAP of each group was above the study targets (79 mm Hg in high target/intervention arm and 72 mm Hg in low target/control arm). As a result, these outcomes may not be truly reflective of the benefits or harms of targeting an MAP above or below 65 mm Hg. The lack of preestablished control and intervention MAP target for eligible studies is a weakness, although this could not be established without prior examination of all the available literature. Additionally, the small participant numbers in the subgroup analyses and the fact that they are likely underpowered are a cause for increased uncertainty in the findings, although subgroup results should mostly be hypothesis-generating. Furthermore, the heterogeneity of the underlying patient populations in the pooled studies (i.e., distributive vs hemorrhagic vs cardiogenic shock) and differences in intervention limits the certainty and generalizability of our findings. Subgroup and sensitivity analyses were conducted in an effort to mitigate this limitation. There was largely no effect modification based on type of shock or control MAP targets for the outcomes of mortality, need for renal replacement therapy, ICU and hospital length of stay, volume of fluids administered, or any adverse events. This may support the generalizability of our findings. However, further research is certainly required to draw firm conclusions on the impact of differing MAP targets on each specific population as subgroup analysis is hypothesis-generating. Finally, the small number of eligible studies did not allow for completion of all subgroup analyses and, thus, limits the ability to infer conclusions in these specific populations. Finally, although we were able to compare the conventionally accepted MAP target of 65 mm Hg to higher targets, there were limited studies comparing whether permissive hypotension results in any difference in outcomes. Future studies on permissive hypotension may allow for earlier deresuscitation of patients.

CONCLUSIONS

In conclusion, our systematic review and meta-analysis demonstrated with moderate certainty that there is no difference in mortality when a higher MAP is targeted in critically ill adult patients with shock. Further studies are needed to determine the impact of MAP on need for renal replacement therapy in this population. Finally, studies to examine subgroup effects as well as permissive hypotension to an MAP of 60 mm Hg would contribute valuable knowledge to the care of this population.

REFERENCES

1. Finfer SR, Vincent J-L, De D: Critical care medicine circulatory shock. N Engl J Med. 2013; 18:1726–1760
2. Evans L, Rhodes A, Alhazzani W, et al.: Surviving sepsis campaign: International guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021; 37:62
3. Boyd JH, Forbes J, Nakada TA, et al.: Fluid resuscitation in septic shock: A positive fluid balance and elevated central venous pressure are associated with increased mortality. Crit Care Med. 2011; 39:259–265
4. McIntyre WF, Um KJ, Alhazzani W, et al.: Association of vasopressin plus catecholamine vasopressors vs catecholamines alone with atrial fibrillation in patients with distributive shock a systematic review and meta-Analysis. JAMA. 2018; 319:1889–1900
5. D’Aragon F, Belley-Cote EP, Meade MO, et al.; Canadian Critical Care Trials Group: Blood pressure targets for vasopressor therapy: A systematic review. Shock. 2015; 43:530–539
6. Hylands M, Moller MH, Asfar P, et al.: Une revue systématique des cibles de tension artérielle sous vasopresseurs chez des adultes gravement malades atteints d’hypotension. Can J Anesth. 2017; 64:703–715
7. Lamontagne F, Richards-Belle A, Thomas K, et al.: Effect of reduced exposure to vasopressors on 90-day mortality in older critically ill patients with vasodilatory hypotension: A randomized clinical trial. JAMA. 2020; 323:938–949
8. Ameloot K, Jakkula P, Hästbacka J, et al.: Optimum blood pressure in patients with shock after acute myocardial infarction and cardiac arrest. J Am Coll Cardiol. 2020; 76:812–824
9. Grand J, Meyer AS, Kjaergaard J, et al.: A randomised double-blind pilot trial comparing a mean arterial pressure target of 65 mm Hg versus 72 mm Hg after out-of-hospital cardiac arrest. Eur Hear Journal Acute Cardiovasc Care. 2020; 9(4_suppl):S100–S109
10. Higgins JPT, Savović J, Page MJ, et al.: RoB 2: A revised Cochrane risk-of-bias tool for randomized trials. BMJ. 2019; 366:l4898
11. Review Manager (RevMan) [Computer program]. Version 5.3, The Cochrane Collaboration, 2014
12. DerSimonian R, Laird N: Meta-analysis in clinical trials revisited. Contemp Clin Trials. 2015; 45:139–145
13. Irwig L, Macaskill P, Berry G, et al.: Bias in meta-analysis detected by a simple, graphical test. Graphical test is itself biased. BMJ. 1998; 316:629–634
14. Turner RM, Bird SM, Higgins JPT: The impact of study size on meta-analyses: Examination of underpowered studies in Cochrane reviews. PLoS One. 2013; 8:e592021–e592028
15. Wetterslev J, Thorlund K, Brok J, et al.: Estimating required information size by quantifying diversity in random-effects model meta-analyses. BMC Med Res Methodol. 2009; 9:1–12
16. Wetterslev J, Jakobsen JC, Gluud C: Trial sequential analysis in systematic reviews with meta-analysis. BMC Med Res Methodol. 2017; 17:1–18
17. Higgins JPT, Thompson SG, Deeks JJ, et al.: Measuring inconsistency in meta-analyses. Br Med J. 2003; 327:557–560
18. Carayannopoulos KL, Lewis K, Alhazzani W. Mean arterial pressure targets and patient-important outcomes in critically ill adults: A systematic review and meta-analysis of randomized trials. PROSPERO - international prospective register of systematic reviews. Available at: https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=219971. Accessed November 7, 2022
19. Guyatt GH, Oxman AD, Vist GE, et al.: GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. Chinese J Evidence-Based Med. 2009; 9:8–11
20. GRADEpro GDT. 2021. Available at: https://www.gradepro.org/. Accessed June 8, 2021
21. Siemieniuk R, Guyatt G: What is GRADE? | BMJ Best Practice. Available at: https://bestpractice.bmj.com/info/toolkit/learn-ebm/what-is-grade/. Accessed July 1, 2021
22. Asfar P, Meziani F, Hamel J-F, et al.; SEPSISPAM Investigators: High versus low blood-pressure target in patients with septic shock. N Engl J Med. 2014; 370:1583–1593
23. Carrick MM, Morrison CA, Tapia NM, et al.: Intraoperative hypotensive resuscitation for patients undergoing laparotomy or thoracotomy for trauma: Early termination of a randomized prospective clinical trial. J Trauma Acute Care Surg. 2016; 80:886–896
24. Lamontagne F, Meade MO, Hébert PC, et al.; Canadian Critical Care Trials Group: Higher versus lower blood pressure targets for vasopressor therapy in shock: A multicentre pilot randomized controlled trial. Intensive Care Med. 2016; 42:542–550
25. Lamontagne F, Day AG, Meade MO, et al.: Pooled analysis of higher versus lower blood pressure targets for vasopressor therapy septic and vasodilatory shock. Intensive Care Med. 2018; 44:12–21
26. Bourgoin A, Leone M, Delmas A, et al.: Increasing mean arterial pressure in patients with septic shock: Effects on oxygen variables and renal function. Crit Care Med. 2005; 33:780–786
27. LeDoux D, Astiz ME, Carpati CM, et al.: Effects of perfusion pressure on tissue perfusion in septic shock. Crit Care Med. 2000; 28:2729–2732
28. Badin J, Boulain T, Ehrmann S, et al.: Relation between mean arterial pressure and renal function in the early phase of shock: A prospective, explorative cohort study. Crit Care. 2011; 15:R135
Keywords:

mean arterial pressure; sepsis; shock; vasoactive agents

Supplemental Digital Content

Copyright © 2022 by the Society of Critical Care Medicine and Wolters Kluwer Health, Inc. All Rights Reserved.