In this prospective, single-center, observational study, we examined plasma levels of ACE2 in patients with sepsis admitted to the ED. We found that plasma ACE2 was associated with patient comorbidity but not with 28-day mortality. In the receiver operating characteristics (ROC) analysis, ACE2 levels did not predict 28-day mortality, in contrast to the widely used laboratory biomarkers CRP, creatinine. Whereas plasma ACE2 did not predict mortality, CRP and creatinine were only weak predictors (AUC < 0.6), as were vital signs (AUC 0.57–0.67), underlining the complexity of risk prediction in sepsis patients.

To our knowledge there are no other studies investigating plasma ACE2 levels in patients with sepsis admitted to the ED. The results of the present study therefore add to the discussion of ACE2’s role in infectious diseases and as a biomarker predicting mortality^6,9,16. In the present study, we confirm findings from other studies in the general population and cohorts of COVID-19 patients which have shown that plasma ACE2 levels increase with age, are higher in men, and in patients with hypertension, congestive heart failure, diabetes, obesity and cancer^9,17. In relation to what has been found in studies on COVID-19 patients, we could not confirm that plasma ACE2 predicts organ failure^9,16,18. In line with other studies^10,17,19,20, plasma ACE2 levels reflected male sex, age and comorbidities also in the present study sample of sepsis patients. Interestingly, ACE2 has been proposed as a biomarker of frailty due to its close association with higher BMI, worse physical function and increased dependency²¹. We did not analyze frailty, but it is well accepted that frailty is associated with increased comorbidity, particularly involving chronic diseases. Thus, a partial explanation for our finding could be that plasma ACE2 in the study subjects reflects a higher level of comorbidity and frailty. We also found an association between plasma ACE2 and admission to ward and IMCU indicating an association with disease severity, although only when the analyses were not adjusted for age and comorbidities.

Although there has also been some discrepancy as regards the prognostic value of ACE2 in predicting COVID-19 outcome^16,22, there is a strong rationale for using ACE2 as a marker in COVID-19. Given that SARS-COV-2 utilizes the membrane-bound ACE2 as a receptor and induces its shedding by activating other membrane-bound enzymes such as ADAM17 and ADAM10⁴, it is plausible that circulating ACE2 levels in COVID-19 patients to a greater extent reflects disease severity, compared to sepsis, based on the findings of the present study. Strengthening this argument, patients with COVID-19 had higher circulating ACE2 levels compared to patients with influenza and similar disease severity¹⁸.

In patients with sepsis, the use of ACE2 as a potential therapy is controversial as ACE2 has been suggested to represent a double-edged sword⁶. On the one hand, high ACE2 activity can protect against both sepsis-induced left ventricular dysfunction, lung inflammation, ARDS, early-stage acute kidney injury (AKI), and mortality by increasing ANG 1–7 and decreasing ANGII activity^6,23,24,25. On the other hand, increased ACE/ANG II is crucial for survival in septic shock, by increasing the vascular resistance and elevating blood pressure⁶, why high ACE2 levels theoretically could increase disease severity by decreasing the natural RAS response to septic shock. Therefore, the balance of the RAS may be of great importance in care of patients exposed to sepsis and septic shock, where ACE2 activity may promote tissue protection and survival under certain conditions, but negatively impact tissue protection and survival in other conditions, by decreasing blood pressure, tissue perfusion, thereby leading to organ failure⁶. The present study does not provide evidence indicating benefits from using ACE2 as a potential therapy in patients with sepsis. It is possible that future studies investigating the longitudinal dynamics of ACE2 in sepsis patients could provide more information.

There is a strong rational for increasing the understanding of how RAS activation can be utilized in the care of sepsis patients. In studies of mice and sheep, the administration of ANG 1–7 reduces LPS/septic shock induced inflammation and attenuates organ injury^26,27. There are concerns that ANG 1–7 administration could worsen septic shock by causing vasodilatation, however, such effects were not observed in the septic shock sheep model²⁷. At the other end of the spectrum, several post-hoc studies based on the Angiotensin II for the Treatment of High Output Shock (ATHOS-3) trial have found that ANGII administration can provide potential benefit on survival on subsets of patients with vasodilatory shock²⁸. Administration of supraphysiological levels of ANGII, or inhibiting ACE2 activity in swine leads to severe lung and kidney injury, reduced blood oxygenation and hypercoagulopathy²⁹.

Theoretically, the timing of ANGII and/or ACE2/ANG1–7 administration to patients with sepsis could be of great importance. In the present study we only have data on ACE2 levels at admission, which does not seem sensitive enough to identify patients at risk of severe outcome. Future studies should include repeated blood sampling, evaluating the dynamics of ACE2 and RAS during disease progression, as a possible key to identifying patients at risk, and finding eventual therapeutic time windows. That the timing of potential modulation of the RAS may be crucial was further demonstrated in a recent post-hoc analysis showing that patients in vasodilatory shock administered ANGII when the norepinephrine-equivalent dose was ≤ 0.25 µg/kg/min had higher survival rates compared to placebo or when the norepinephrine-equivalent dose was > 0.25²⁸.

Strengths and limitations

One important strength of the present prospective observational cohort study is the large study sample. Also, all patient records in this study were thoroughly revised by infectious disease physicians to assure correct diagnoses. This study, however, has several limitations. Patients were only enrolled during office hours, implicating potential selection bias. Also, only admission samples were available, which excludes analyses of dynamic changes. Studies on severe COVID-19 patients indicate that a dramatic increase in circulating ACE2 levels is observed first a few days after admission. Reindl-Schwaighofer et al.¹⁸ found that circulating ACE2 levels in patients with severe COVID-19 increased sevenfold from early to late following admission, and that this increase was accompanied by increased ANG1–7 levels and ANG1–7/ANGII ratio, suggesting a shift towards the alternative RAS and increased organ protection later in the disease course. Finally, no healthy controls were used, why we cannot say anything about how the plasma ACE2 levels in sepsis patients relate to the general population. The study data was gathered when the Sepsis-2 criteria defined sepsis³⁰, which is no longer the current definition.

In the present study, we included all patients presenting with symptoms reaching the sepsis-2 definition. This could be seen as a limitation since an unselected sample of sepsis patients were included. The site of infection underlying sepsis could be more or less predictive of mortality. However, upon admission to the ED, the underlying cause of sepsis is not always clear. To account for differences in the underlying cause of sepsis, we included infection site in the multivariable analyses when significant differences were observed between survivors and non-survivors.

In conclusion, in patients with sepsis, plasma ACE2 levels are increased in those with male sex, high age and comorbidities, including diabetes, obesity, cardiovascular and cancer diseases. ACE2 levels were not associated with 28-mortality, nor with organ failure. Contrary to our hypothesis and to animal studies, we did not find ACE2 to be a promising marker in sepsis. To our knowledge, this is one of the first studies elucidating on plasma ACE2 in patients with sepsis admitted to the ED. Future studies which investigate the dynamics of circulating ACE2 levels in longitudinal studies with repeated sampling are warranted. An increased understanding of the dynamics of RAS dysfunction in sepsis patients could potentially help generate individualized treatment options, especially in those patients who progress into vasodilatory shock.