Elsevier

Journal of Critical Care

Volume 29, Issue 1, February 2014, Pages 157-160
Journal of Critical Care

Clinical Potpourri
Does high-dose vasopressor therapy in medical intensive care patients indicate what we already suspect?

https://doi.org/10.1016/j.jcrc.2013.09.004Get rights and content

Abstract

Purpose

This study was conducted to determine the association between vasopressor requirement and outcome in medical intensive care patients in an environment where treatment is not withdrawn.

Materials and Methods

This was an observational study of patients in the medical intensive care unit (ICU) over a period of 18 months to determine the correlation between vasopressor requirement and mortality. Outcome was determined for all medical ICU patients, for patients receiving “low dose” (< 40 μg/min) vasopressors (noradrenaline and/or adrenaline) or “high dose” (≥ 40 μg/min) vasopressors. Receiver operator characteristic curves were constructed for ICU and hospital mortality and high-dose vasopressor use. High-dose vasopressor use as an independent predictor for ICU and hospital mortality was also determined by multiple logistic regression analysis.

Results

Patients receiving high-dose noradrenaline at any time during their ICU admission had an 84.3% mortality in ICU and 90% in hospital. The receiver operator characteristic curves for high-dose vasopressors had an area under the curve of 0.799 for ICU mortality and 0.779 for hospital mortality. High-dose vasopressor was an independent predictor of ICU mortality, with an odds ratio of 5.1 (confidence interval, 2.02-12.9; P = .001), and of hospital mortality, with an odds ratio of 3.82 (confidence interval 1.28-11.37; P = .016).

Conclusions

The requirement for high-dose vasopressor therapy at any time during ICU admission was associated with a very high mortality rate in the ICU and the hospital.

Introduction

As practicing intensivists, we intuitively appreciate when a patient's situation is becoming dire. We spend time at the bedside, examine the patient, order tests, and view the trend in the physiological parameters. In this setting, we may observe the inexorable escalation of vasopressor support in an attempt to maintain organ perfusion [1]. Vasopressor therapy is commonly used to defend organ perfusion when there is an inadequate response to intravascular volume repletion in patients in the intensive care unit (ICU). Indeed vasopressor therapy is recommended for septic shock, in particular [2], [3], and it has also been shown to favorably influence outcome in patients with septic shock [4]. However, guidelines are lacking as to the maximal dose recommended for subgroups of critically ill patients.

Given the high incidence of shock and septic shock, in particular, presenting to ICUs [5], vasopressor therapy is very commonly used. However, at some point and in some patients, it becomes evident that there is poor or no response to vasopressor therapy, such as high-dose adrenaline, noradrenaline, and/or vasopressin. Most intensivists recognize this scenario as one in which the patient is unlikely to do well and indeed may make the decision to withdraw therapy on the grounds of medical futility. The literature is scant in terms of the prognostic predictive value of high-dose vasopressor therapy in this setting [6], [7]. This issue is somewhat obscured by the concept of the self-fulfilling prophesy represented by withdrawal of treatment deemed to be futile (ie, “non-response to high-dose vasopressor therapy is a good indicator that the patient is not able to survive the current illness,” so treatment is withdrawn and the patient dies).

We were interested in what actually happens to patients receiving high-dose vasopressor therapy in the clinical setting where treatment withdrawal is not practiced. The authors work in a cultural setting where withdrawal of treatment is not carried out, regardless of the perceived futility of ongoing intensive care treatment. The outcome of patients, in terms of mortality, is therefore not affected by the practice of withdrawal of therapy and represents a “true” outcome.

We undertook an observational study over the period April 27, 2008, to August 3, 2010, examining the outcome of patients admitted to our ICU who received no vasopressors, lower-dose vasopressors, or high-dose vasopressors.

Section snippets

Materials and methods

We applied to and received from the institutional ethics committee a waiver of the requirement for informed consent (review of records only and use of non-identified patient data). Data were then obtained from our unit database on all patients admitted over the period April 27, 2008, to August 3, 2010, representing 917 patients. Detailed data were then collected for patients receiving noradrenaline or adrenaline, being the vasopressors most commonly used in our unit. Of the 917 patients

Results

The results shown in Table 1 indicate a significant difference (P < .05) between the low- and high-dose vasopressor groups for age (younger patients received high dose) and APACHE II score (sicker patients in the high-dose group). The high-dose vasopressor group also had more patients with hematologic malignancies, but with less preexisting cancer. High-dose vasopressor therapy was also associated with more vasopressin use, not surprising given that vasopressin is used for refractory

Discussion

We found that the use of high-dose vasopressor therapy in the medical ICU (MICU) is associated with a very high mortality of 84.3% in the ICU and 90% for the hospital admission in our population of adult medical intensive care patients. We regard this as “true outcome” because we do not practice withdrawal of life support drugs and interventions. Our outcome data for patients requiring high-dose vasopressors represent a 4-fold increase in mortality over the general MICU population during the

Conclusion

We have shown that the requirement for high-dose vasopressor therapy is a predictor for poor outcome, and more than 90% of patients in our MICU who received more than 40 μg/min of vasopressors at any time during the ICU admission died. Guidelines are lacking for the acceptable range of high-dose vasopressors and inotropes in specific patient subgroups.

References (9)

There are more references available in the full text version of this article.

Cited by (30)

  • The Relationship Between Norepinephrine Equivalent Dose of Vasopressors Within 24 Hours From the Onset of Septic Shock and In-Hospital Mortality Rate

    2023, Chest
    Citation Excerpt :

    In the present study, the in-hospital mortality rate dramatically and continuously increased as the maximum NEE dose in the first 24 h of shock onset increased, even after adjustment for the patient baseline characteristics and the severity score. Although the relationships between high-dose vasopressors and death have been described previously in smaller cohorts,4-6 our study is the largest cohort of patients with septic shock who condition required a maximum NEE of ≥ 1.0 μg/kg/min in the first 24 h after shock onset. Our study has two important implications.

  • Hitting the Vasopressor Ceiling: Finding Norepinephrine Associated Mortality in the Critically Ill

    2021, Journal of Surgical Research
    Citation Excerpt :

    One observational cohort study demonstrated a positive correlation between number of vasopressors required and mortality rate, with an observed mortality rate of 92% in the setting of three vasopressors at maximum infusion.8 Another retrospective study showed that patients who received more than 40 mcg/min of any vasopressor suffered a 90% in-hospital mortality rate.12 In spite of the evidence that high vasopressor doses correlate with mortality, there is currently no guidance regarding what norepinephrine dose correlates with clinical futility.

  • Methylene Blue for Vasoplegia During Extracorporeal Membrane Oxygenation Support

    2021, Journal of Cardiothoracic and Vascular Anesthesia
  • Angiotensin II in Vasodilatory Shock

    2019, Critical Care Clinics
    Citation Excerpt :

    By one estimate, only 17% of patients with septic shock requiring vasopressor therapy of greater than or equal to 1 mcg/kg/min of norepinephrine-equivalent dosing survive to 90 days.12 In addition, high-dose catecholamine therapy has been shown to be independently predictive of mortality after controlling for many factors, including severity of illness.33 Catecholamine monotherapy is also associated with significant cardiac side effects, morbidity, and mortality, an effect that is correlated with the cumulative dose of catecholamines, the number of different catecholamines used, and the duration of therapy.34

  • Regional differences in the treatment of refractory vasodilatory shock using Angiotensin II in High Output Shock (ATHOS-3) data

    2019, Journal of Critical Care
    Citation Excerpt :

    As with the case of combination vasopressors, the guidelines remain unclear on timing of steroid initiation, in relation to vasopressor dosing. Part of this issue is the lack of clarity on what constitutes catecholamine-resistant hypotension [2,28,33-35]. The ATHOS-3 trial designated a NED of >0.2 μg/kg/min based on a calculated 50% mortality using SOFA scores, but other higher doses that cited nearly 80–100% mortality have been proposed as well [33,34,36].

  • Intensity of Vasopressor Therapy for Septic Shock and the Risk of In-Hospital Death

    2017, Journal of Pain and Symptom Management
    Citation Excerpt :

    Given the high mortality associated with septic shock, a method of distinguishing patients who do or do not have a reasonable chance of surviving with aggressive treatment could help clinicians and families weigh the options and make informed decisions. Anecdotal observations at our institution and a few published reports11–13 have suggested that the chance of surviving septic shock decreases as the intensity of vasopressor therapy needed to treat the condition escalates. The present observational cohort study examined in-hospital mortality as a function of therapeutic intensity by analyzing vasopressor use and survival status in a consecutive series of adult patients with septic shock who were admitted to Winthrop University Hospital over a four-year period.

View all citing articles on Scopus

Conflict of interest: None of the authors have any conflict of interest in the conduct or reporting of the research described in this manuscript.

View full text