Highlights
– In TARTARE-2S (n=219), a protocol targeting tissue perfusion (capillary refill time, peripheral temperature, lactate) and allowing MAP 50–65 mm Hg produced no improvement in days alive with normalized lactate and vasopressor-free at 30 days compared with guideline-based MAP-guided standard care.
– Median days alive and free of vasopressors with normal lactate were 23 versus 22 days (difference 0.59 days; 95% CI, −3 to 4), and 30‑day mortality was similar (24.7% vs 27.8%).
– The TTP approach resulted in lower achieved MAP but did not raise safety signals; implications concern tailoring vasopressor targets and the challenge of translating microcirculatory endpoints into clinical benefit.
Background
Septic shock is a major cause of intensive care morbidity and mortality worldwide. Standard resuscitation focuses on macrocirculatory goals—restoring mean arterial pressure (MAP), cardiac output, and oxygen delivery—to reverse tissue hypoperfusion. Current guidelines (Surviving Sepsis Campaign) recommend targeting a MAP of at least 65 mm Hg in most patients, but several physiologic and clinical observations highlight that macrocirculatory variables may not reliably reflect microcirculatory perfusion. Concerns about vasopressor-associated harm, particularly at higher MAP targets, have motivated research into individualized strategies that prioritize tissue perfusion markers (e.g., capillary refill time, peripheral temperature, lactate clearance) and may permit lower vasopressor exposure.
Study design
TARTARE-2S was a randomized, parallel-group, open-label trial enrolling adult ICU patients with septic shock and arterial lactate >3 mmol/L across three European university hospital ICUs (2016–2022). A total of 219 patients were randomized to either targeted tissue perfusion (TTP; n=111) or mean arterial pressure–guided standard care (SC; n=108). Allocation was stratified by center and presence of chronic arterial hypertension.
The TTP protocol used bedside indices of peripheral perfusion (capillary refill time and peripheral skin temperature), arterial lactate concentrations, and a permissive MAP target of 50–65 mm Hg. The SC group received hemodynamic targets guided by the 2012 Surviving Sepsis Campaign recommendations (which emphasize MAP ≥65 mm Hg among other goals).
The primary outcome was days alive in 30 days with normal lactate and without vasopressor or inotropic support (a composite intended to reflect clinically meaningful recovery of perfusion without ongoing vasopressor dependence). Secondary outcomes included components of the primary endpoint (time to vasopressor discontinuation, lactate normalization), days alive without organ support, 30‑day mortality, and serious adverse reactions.
Key findings
Of the randomized cohort, 97 (87.4%) in the TTP group and 97 (89.8%) in the SC group (total n=194) were included in the primary analysis. The primary outcome—the median days alive in 30 days with normal lactate and without vasopressors/inotropes—was 23 (IQR 10–27) in the TTP group versus 22 (IQR 1–27) in the SC group. The difference in medians was 0.59 days (95% CI, −3 to 4), indicating no statistically or clinically important benefit of TTP over SC for the primary composite outcome.
Secondary outcomes likewise showed no significant differences. Individual components of the composite—time to lactate normalization and time to vasopressor discontinuation—were not meaningfully different between groups. Organ‑support–free days and rates of renal replacement therapy, mechanical ventilation, or vasopressor duration were similar. At 30 days, mortality was 24.7% in the TTP arm and 27.8% in the SC arm (absolute difference −3.1 percentage points), not statistically significant.
Physiologic measures aligned with the study design: achieved MAP values were lower in the TTP group compared with SC, consistent with the permissive lower MAP target. Importantly, no excess of serious adverse events or safety signals was observed in the lower‑MAP TTP arm.
Effect sizes and precision
The primary outcome median difference was small (0.59 day) and the 95% confidence interval crossed zero (−3 to 4 days), indicating the trial did not demonstrate superiority and cannot exclude a small benefit or harm within that range. Mortality differences were also modest and statistically non‑significant.
Expert commentary and interpretation
TARTARE-2S addresses a clinically important question: can resuscitation aimed at restoring tissue perfusion while accepting lower MAP reduce vasopressor exposure without compromising outcomes? The trial’s negative result—that TTP did not improve days alive and off vasopressors with normalized lactate—has several plausible explanations and important implications.
Context with existing evidence
The findings echo aspects of prior trials comparing higher versus lower MAP targets in septic shock. For example, the SEPSISPAM trial (Asfar et al., NEJM 2014) randomized patients to higher (80–85 mm Hg) versus lower (65–70 mm Hg) MAP targets and found no overall mortality difference, though some renal outcomes differed in patients with chronic hypertension. TARTARE-2S differs by integrating peripheral perfusion measures and accepting even lower MAP ranges (50–65 mm Hg) when tissue perfusion appeared adequate.
Physiologically, it is established that macrocirculatory restoration does not guarantee microcirculatory recovery: the septic microcirculation may remain heterogeneous despite normalized blood pressure or cardiac output. This observation motivated interest in peripheral perfusion–guided resuscitation. Yet translating improved bedside perfusion indices into robust patient-centered outcomes remains challenging.
Strengths
The trial randomized a pragmatic ICU population with an objective lactate inclusion threshold (>3 mmol/L), used a prespecified composite outcome linking physiologic recovery and vasopressor independence, and achieved biologically meaningful separation in achieved MAP between groups. The multicenter European ICU setting increases external validity for similar healthcare environments.
Limitations
Key limitations temper interpretation. TARTARE-2S was open‑label, which is unavoidable in hemodynamic trials but can introduce performance bias. The composite primary endpoint—days alive with normal lactate and without vasopressors—while clinically sensible, is novel and may be influenced by practice patterns around vasopressor weaning and frequency of lactate measurement. The sample size, although adequate for the chosen primary endpoint, may have been underpowered to detect modest but clinically relevant differences in mortality or organ‑support needs. Subgroup effects (for example in patients with chronic hypertension) may not be excluded and require prespecified power to evaluate. Finally, the trial setting and protocols reflect European ICUs and may not generalize to settings with different staffing, monitoring, or vasopressor practices.
Clinical implications
For practicing clinicians, TARTARE-2S suggests that targeting bedside tissue perfusion while tolerating lower MAPs (50–65 mm Hg) does not clearly improve the composite outcome of perfusion-normalization plus vasopressor-free days over standard MAP-guided care. The absence of increased harm is notable and may support individualized hemodynamic goals in selected patients, particularly when microcirculatory indices indicate sufficient perfusion despite lower MAP. However, broad adoption of permissive low-MAP strategies based solely on these findings is not warranted; decisions should be individualized, considering patient comorbidities (e.g., chronic hypertension, cerebral or coronary disease), and clinical context.
Conclusion
The TARTARE-2S randomized trial found that a tissue‑perfusion‑guided resuscitation strategy allowing lower MAP did not increase days alive with normal lactate and without vasopressor/inotropic drugs at 30 days compared with guideline-directed MAP‑guided care in patients with septic shock and lactate >3 mmol/L. No additional safety concerns emerged. These data add to a growing literature that questions one-size-fits-all MAP targets and underscores the complexity of converting microcirculatory-focused physiologic endpoints into superior patient‑centered outcomes. Future work should further define which subgroups (if any) benefit from perfusion‑guided strategies, refine bedside microcirculatory assessments, and explore optimal approaches to minimize vasopressor exposure without compromising organ perfusion.
Funding and clinicaltrials.gov
Funding details and trial registration information are reported in the original manuscript: Pettilä V et al., Crit Care Med. 2025 (doi: 10.1097/CCM.0000000000006899) . Refer to the published paper for exact funding sources and registration identifiers.
References
1. Pettilä V, Pfortmüller CA, Perner A, Merz TM, Wilkman E, Hästbacka J, Lang MF, Lombardo P, Selander T, Jakob SM, Takala J. Targeted Tissue Perfusion Versus Macrocirculatory-Guided Standard Care in Patients With Septic Shock: A Randomized Clinical Trial—The TARTARE-2S Trial. Crit Care Med. 2025 Oct 17. doi: 10.1097/CCM.0000000000006899 . Epub ahead of print. PMID: 41105050 .
2. Asfar P, Meziani F, Hamel JF, et al.; SEPSISPAM Investigators. High versus low blood-pressure target in patients with septic shock. N Engl J Med. 2014 Oct 9;371(17):1573–1584. doi:10.1056/NEJMoa1404712 .
3. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001 Nov 8;345(19):1368–1377. doi:10.1056/NEJMoa010307 .
4. Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis‑3). JAMA. 2016 Feb 23;315(8):801–810. doi:10.1001/jama.2016.0287 .
5. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock. (See the most recent guideline updates for contemporary recommendations; earlier bundles informed the comparator protocol used in TARTARE‑2S.)