TYPE Original Research

PUBLISHED 28 October 2022

DOI 10.3389/fmed.2022.1014276

Methylene blue dosing

OPEN ACCESS strategies in critically ill adults
with shock—A retrospective
EDITED BY
Pavel Michalek,
General University Hospital in
Prague, Czechia

REVIEWED BY
cohort study
Driss Laghlam,
Clinique Ambroise Paré, France
Mathieu Jozwiak, Sibel Sari-Yavuz1 , Ka-Lin Heck-Swain1 , Marius Keller1 ,
Center Hospitalier Universitaire de
Nice, France Harry Magunia1 , You-Shan Feng2 , Helene A. Haeberle1 ,
*CORRESPONDENCE Petra Wied1 , Christian Schlensak3 , Peter Rosenberger1 and
Michael Koeppen
[email protected]
Michael Koeppen1*
1
SPECIALTY SECTION Department of Anesthesiology and Intensive Care Medicine, University Hospital Tübingen,
This article was submitted to Tübingen, Germany, 2 Institute for Clinical Epidemiology and Applied Biostatistics (IKEaB),
Intensive Care Medicine and Eberhard-Karls-University Tübingen, Tübingen, Germany, 3 Department of Thoracic and
Anesthesiology, Cardiovascular Surgery, University Hospital Tübingen, Tübingen, Germany
a section of the journal
Frontiers in Medicine

RECEIVED 08 August 2022 Background: Shock increases mortality in the critically ill and the mainstay of
ACCEPTED 07 October 2022 therapy is the administration of vasopressor agents to achieve hemodynamic
PUBLISHED 28 October 2022
targets. In the past, studies have found that the NO-pathway antagonist
CITATION
Sari-Yavuz S, Heck-Swain K-L, Keller M, methylene blue improves hemodynamics. However, the optimal dosing
Magunia H, Feng Y-S, Haeberle HA, strategy remains elusive. Therefore, we investigated the hemodynamic and ICU
Wied P, Schlensak C, Rosenberger P
outcome parameters of three different dosing strategies for methylene blue.
and Koeppen M (2022) Methylene blue
dosing strategies in critically ill adults Methods: We performed a retrospective cohort study of patients in shock
with shock—A retrospective cohort
study. Front. Med. 9:1014276.
treated with methylene blue. Shock was defined as norepinephrine dose >0.1
doi: 10.3389/fmed.2022.1014276 µg/kg/min and serum lactate level >2 mmol/l at the start of methylene
COPYRIGHT blue administration. Different demographic variables, ICU treatment, and
© 2022 Sari-Yavuz, Heck-Swain, Keller, outcome parameters were evaluated. To compare the differences in the
Magunia, Feng, Haeberle, Wied,
Schlensak, Rosenberger and Koeppen.
administration of vasopressors or inotropes, the vasoactive inotropic score
This is an open-access article (VIS) was calculated at different time points after starting the administration
distributed under the terms of the of methylene blue. Response to methylene blue or mortality at 28 days
Creative Commons Attribution License
(CC BY). The use, distribution or were assessed.
reproduction in other forums is
Results: 262 patients from July 2014 to October 2019 received methylene
permitted, provided the original
author(s) and the copyright owner(s) blue. 209 patients met the inclusion criteria. Three different dosing strategies
are credited and that the original were identified: bolus injection followed by continuous infusion (n = 111),
publication in this journal is cited, in
accordance with accepted academic bolus injection only (no continuous infusion; n = 59) or continuous infusion
practice. No use, distribution or only (no bolus prior; n = 39). The groups did not differ in demographics,
reproduction is permitted which does
ICU scoring system, or comorbidities. In all groups, VIS decreased over
not comply with these terms.
time, indicating improved hemodynamics. Cardiogenic shock and higher
doses of norepinephrine increased the chance of responding to methylene
blue, while bolus only decreased the chance of responding to methylene
blue treatment. 28-day mortality increased with higher SAPSII scores and
higher serum lactate levels, while bolus injection followed by continuous
infusion decreased 28-day mortality. No severe side effects were noted.

Frontiers in Medicine 01 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

Conclusion: In this cohort, methylene blue as a bolus injection followed by

continuous infusion was associated with a reduced 28-day mortality in patients
with shock. Prospective studies are needed to systematically evaluate the role
of methylene blue in the treatment of shock.

KEYWORDS

shock, methylene blue, critical care, retrospective study, clinical trial

Background mainly due to the complex pharmacological properties of

methylene blue. It has a large volume of distribution (up to 250 l)
Shock results from a multitude of pathophysiological and a long half-life of up to 24 h (13), making the prediction of
processes and leads to hemodynamic collapse and death clinical effects and hemodynamic responses difficult.
if left untreated. Clinicians differentiate forms of shock At our institution, the administration of methylene blue
such as cardiogenic, hypovolemic, obstructive, or distribution belongs to the standard operating procedure in the treatment
shock (including septic shock) (1). They all share the of refractory shock. Three different dosing strategies have been
same features, leading to hypotension and impaired oxygen used: bolus of methylene blue followed by continuous infusion,
delivery or utilization in the peripheral circulation (2), either bolus injection only, or continuous infusion only. In the present
due to loss of vascular tone or decreased cardiac output. retrospective cohort analysis, we performed an unbiased analysis
This jumpstarts a vicious circle consisting of an altered of these three different regimens and investigated whether one
microcirculation leading to the production and release of dosing strategy leads to improved hemodynamic stability over
pro-inflammatory mediators such as tumor necrosis factor the other. Next, we searched for variables that correlate with a
alpha (TNFa) or interleukin-6 (IL-6), which in turn induce clinical response to methylene blue, defined as a reduction in
the expression of the inducible NO-synthase (iNOS) with vasoactive or inotropic substances. Finally, we assessed which
subsequent production of the vasodilator nitric oxide (NO). NO clinical variables correlate with reduced mortality at 28 days in
acts as a vasodilator that directly counteracts the endogenous critically ill adults.
or exogenous vasoconstrictors and thus decreases the vascular
tone (3). Interestingly, increased endothelial NO production
Methods
contributes substantially to the pathophysiology of shock (4, 5)
and leads to a decrease in mean arterial pressure resulting Study population and ethics
in impaired organ perfusion. If volume resuscitation alone
cannot reestablish target values, patients require vasopressors. All data were retrospectively collected from medical records
In recent decades, different vasopressors, such as dopamine, from the University Hospital of Tuebingen. The study was
epinephrine, or norepinephrine, have been clinically tested in approved by the Ethics Committee of the University Hospital
shock patients (6–8). Norepinephrine has emerged as the first- Tuebingen (IRB# 768/2018BO2), which waived the need for
line vasopressor agent in most forms of shock. In patients with informed consent because patient anonymity was maintained.
severe circulatory failure, additional vasoactive substances are All methods were approved by the local IRB and performed
required to stabilize the blood pressure. Several clinical studies in accordance with the Declaration of Helsinki and the
have evaluated NO-synthase inhibitors in experimental and relevant guidelines.
clinical shock treatment (9), but the use of these substances never
took hold in clinical therapy.
In 1992, patients with septic shock (a form of distributive Methylene blue therapy
shock) received the first fully synthetic drug ever used in
medicine invented in 1876 (10): methylene blue. Methylene blue In our institution, patients received methylene blue
was clinically tested as a shock treatment in various clinical therapy according to the standard operating procedure (SOP).
trials (11, 12) evaluating different dosing strategies for methylene At the discretion of the attending physician, methylene
blue dosing regimens. They range from bolus administration blue can be considered if norepinephrine dose increases
regimens to bolus administration followed by continuous >0.1 µg/kg/min. When norepinephrine administration does
infusion, or continuous infusion only without additional bolus not establish sufficient hemodynamics (norepinephrine >0.3
(13). To date, no studies have investigated the optimal dose µg/kg/min), vasopressin at a dose of 0.06 IU/kg/min is added
strategy for methylene blue in critically ill patients with shock, for vasopressor therapy. If mean arterial pressure cannot be

Frontiers in Medicine 02 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

maintained at more than 65 mmHg, methylene blue is added to Vasoactive inotropic score
the therapy by standard.
Three different dosing strategies have been used in our In the present study, different vasoactive substances were
institution in the past: bolus injection alone, continuous used. Often patients received different substances at once. These
infusion without bolus, injection, or bolus injection followed can be subdivided into two broad categories, inotropes and
by continuous infusion (bolus 2 mg/kg; continuous infusion vasopressor. As inotropes, dobutamine and milrinone were
with 0.25 mg/kg/h). We define a patient as a “methylene blue used. In the category of the vasopressors, norepinephrine
responder” according to the following criteria: VIS decreases by is considered the first-line treatment option; vasopressin is
>10% within 3 h after administration and mean arterial pressure considered additionally. Epinephrine is used in the clinical
>65 mmHg. setting for both indications. Most patients received more
than one substance of the respective category. Furthermore,
vasoactive and inotropic doses can differ greatly between
institutions. To quantify the degree of hemodynamic support
Data collection
objectively, we calculated the so-called vasoactive inotropic score
(VIS) as described (15). The VIS incorporates vasoconstrictor
All patient records who were treated at the Department of
and inoptropes in a single variable. The VIS was calculated using
Anesthesiology and Intensive Care Medicine between July 2014
the following formula:
and October 2019 were retrospectively screened for inclusion
in this observational cohort study. We selected all patients
who received methylene blue during their stay in the ICU.
VIS = Dopamine(µg/kg/min) + Dobutamine(µg/kg/min)
We included patients with the following inclusion criteria:
Norepinephrine dose 0.1 µg/kg/min, serum lactate 2 mmol/l +[100×Epinephrine(µg/kg/min)]
and methylene blue treatment. The following exclusion criteria +[10.000×Vasopressin(U/kg/min)]
were applied: Age < 18 years, no vasopressor or inotropic +[100×Norepinephrine(µg/kg/min)]
therapy, administration of methylene for other reasons than
+[10×Milrinon (µg/kg/min)]. (1)
hemodynamic compromise.
Based on data from the clinical information system,
a database was generated that contains relevant patient
Statistical analysis
information, including age, sex, date of admission and discharge
from the ICU. Furthermore, we recorded ICU prognosis scores,
Data for continuous variables are expressed as the mean
such as Sequential Organ Failure Assessment (SOFA), Simplified
± standard deviation in the case of a normal distribution
Acute Physiology Score II (SAPS II) and Acute Physiology
and as the median (interquartile range) in the case of a
and Chronic Health Evaluation (APACHE) scores. All scores
nonnormal distribution. Variables from the different groups
presented were calculated for the day of methylene blue
were compared using the ANOVA test with Tukey-Kramer
administration. We determined comorbidities from the patient
post-hoc test for multiple comparison. Categorical variables
records as well as the dosage of vasopressor or inotropic therapy.
expressed as proportions were compared using the chi-square
All ICU progress and discharge notes were screened for
test. To assess the association of different variables with the
severe side effect of methylene blue. The most severe form of side
response status of methylene blue or mortality at 28 days,
effects is anaphylaxis, followed by shortness of breath, nausea,
univariate logistic regression was used. Variables with biologic
vomiting. The most common side effect is a bluish to greenish
plausibility to influence the response status to methylene
discoloration of the skin or urine, which is self-resolving (14).
blue or mortality with a significant association in univariate
regression (defined as p < 0.1) were considered for inclusion
in a multivariate logistic regression model. Collinearity was
Classification of shock tested using variance inflation factors and condition indices.
If two or more variables showed collinearity, the variable
Based on the discharge note from the ICU, we retrospectively with the lowest p-value in the univariate logistic regression
classified the patients into a category based on available was included in the multivariate model. Residuals were tested
parameters. Patients with documented infection and shock were for normal distribution. Goodness-of-fit was tested using the
classified as septic shock. Patients with extended hemodynamic Hosmer-Lemeshow test. In the logistic regression analysis, we
monitoring and a cardiac index < 2.5 l/min/m2 were classified performed full-data set analysis. No missing data was noted for
as cardiogenic shock. Patients with a shock criterion as described the variables integrated in the univariate or multivariate logistic
above, but could not be categorized into one of the other two regression. All statistical analyzes for this study were performed
categories, were classified as vasoplegic syndrome. in Prism 9 (GraphPad Software Inc.) and JMP 16.0 (SAS Institute

Frontiers in Medicine 03 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

types of shock did not differ between the three different

dosing strategies.
The most common comorbidity in all data sets was coronary
artery disease (81.8% of total) and arterial hypertension
(80.3% of total). Although there were significant differences
in preexisting lung disease (e.g., chronic obstructive lung
disease, emphysema), all groups had a fairly low number.
In summary, the groups were comparable in their baseline
demographic data.

ICU treatment variables

Next, we investigated whether the different dosing

strategies differed in their disease severity. Interestingly,
28-day mortality was significantly lower in the cohort
treated with bolus injection first with subsequent continuous
infusion (53.1%), compared to the other dosing strategies
(bolus only: 71.2%; continuous infusion only: 74.3%;
Table 2). We also assessed short-term mortality, which
we defined as death within 12 h after administration of
methylene blue. The differences in 28-day and short-term
mortality were not due to differences in disease severity,
as SAPSII, APACHE and SOFA scores were similar
between the groups. However, patients with bolus +
continuous infusion support underwent longer mechanical
ventilator support.
FIGURE 1 Next, we analyzed the differences in the support
Patient selection strategy.
of vasoactive substances. As expected, patients with
bolus + continuous infusion received the highest cumulative
dose of methylene blue. All patients received more than
Inc., Cary, US). The p values are two-tailed and values < 0.05 are one vasoactive substance. In our institution, we regularly
considered statistically significant. use norepinephrine, epinephrine, vasopressin, milrinone
and dobutamine for the treatment of hemodynamic
instability. We found that at the beginning of methylene
Results blue therapy, all patients received a comparable degree
of pharmacological hemodynamic support (Table 2).
Patient selection and characteristics Indicators of decreased perfusion, such as lactate and
blood pH, were not significantly different between
The selection process is depicted in Figure 1. In total, we the groups.
identified 209 patients for analysis.

Demographic data Methylene blue dosing strategy,

hemodynamic parameters, and
In our dataset, we identified three different dosing strategies: vasopressor requirements
53% (n = 111) of the patients received a bolus injection
followed by continuous infusion. 28% of the patients received To understand whether dosing strategy influences the
only a bolus injection (n = 59) and 19% received only hemodynamic response to methylene blue administration, we
a continuous infusion of methylene blue without prior analyzed the basic hemodynamic parameters of mean arterial
bolus (n = 39). Baseline demographic characteristics, such pressure and heart rate (0, 1, 2, and 5 h), after methylene
as age, sex, height, weight, and body mass index, were blue (Figures 2A,B). The hemodynamic response was similar
similar between these groups (Table 1). The distribution across between the three different dosing strategies. In a subset of

Frontiers in Medicine 04 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

TABLE 1 Demographic data and comorbidities in shock patients.

Total Bolus + Bolus only Continuous p-values

(n = 209) infusion (n = 59) infusion only
(n = 111) (n = 39)

Percentage of total 53 28 19
Age–year (mean ± SD) 65.0 ± 13.7 64.5 ± 13.9 67.9 ± 12.6 62.5 ± 14.4 0.1279
Male sex–no. (%) 74.1% 75.7% 72.9% 71.8% 0.8623
Height–cm (mean ± SD) 173.2 ± 9.5 174.3 ± 9.2 172.8 ± 10.0 170.9 ± 9.3 0.1566
Weight–kg (mean ± SD) 84 ± 19 86 ± 21 85 ± 18 79.4 ± 11.8 0.1962
Body Mass Index (kg/m2 ) 28.1 ± 5.7 28.2 ± 6.4 28.4 ± 5.2 27.1 ± 3.5 0.5232
Etiology of shock n (% in group)
Sepsis 82 (39.2) 40 (36.0) 24 (40.7) 18 (46.1) 0.5192
Cardiogenic 64 (30.6) 31 (27.9) 20 (33.9) 13 (33.3) 0.6662
Vasoplegia 63 (30.1) 40 (36.0) 15 (25.4) 8 (20.5) 0.1242
Comorbidities
Lung disease (%) 29 (13.8) 11 (09.0) 13 (24.5) 5 (14.3) 0.0153
Coronary artery disease 171 (81.8) 104 (85.9) 41 (77.3) 26 (74.3) 0.1677
Peripheral artery disease 23 (11.0) 17 (14.0) 4 (7.5) 2 (5.7) 0.1025
Chronic kidney disease 33 (15.8) 14 (11.5) 12 (22.6) 7 (20.0) 0.3949
Arterial hypertension 168 (80.3) 103 (85.1) 39 (73.5) 26 (74.3) 0.0213
Diabetes mellitus 57 (27.2) 36 (29.7) 10 (18.8) 11 (31.4) 0.3468
Autoimmune disease 31 (14.8) 15 (12.3) 8 (15.0) 8 (22.9) 0.6453

patients, systemic vascular resistance (SVR) was measured In summary, all methylene blue treatment regimens are
by pulmonary artery catheterization or transpulmonary equally effective in promoting hemodynamic stabilization.
thermodilution. The evaluation of the SVR measurement closest
to the administration of methylene blue and 12 h later did
not vary between the different dosing strategies. In a subset Response rate to methylene blue
of patients (n = 157), advanced hemodynamic monitoring
(such as pulmonary artery catheter or transpulmonary Methylene blue improves the mean arterial pressure and
thermodilution) was used. Among these patients, cardiac decreases the requirement for vasopressors in critically ill
index was determined. At the administration of methylene patients. However, the response to methylene blue remains
blue, the median cardiac index was 2.97 (interquartile range unpredictable and a proportion of patients remain unresponsive
2–31 – 3.61), which remained essentially unchanged over 12 h to administration. We investigated whether the response to
after methylene blue administration (2.86; interquartile range methylene blue varies between different treatment strategies. To
2.14–3.52). Furthermore, the cardiac indices were not different do this, we analyzed the proportion of clinical responders within
between the different treatment regimens (0 h: p = 0.0714; 12 h the respective group (Figure 3A). In the total cohort, 59.2%
p = 0.1629). of the patients responded to methylene blue, with comparable
Moreover, we investigated whether the vasopressor percentages found in the bolus + continuous infusion and bolus
requirements changed differently over time according groups (Figures 3B–D). On the contrary, continuous infusion
to the dosing. As shown in Figures 2C,D the doses of yielded a lower percentage of responders (44.7%, Figure 3E). In
norepinephrine (used in 100% of patients) or VIS decreased a nominal logistic regression analysis, the cumulative dose of
over time in response to the administration of methylene methylene blue was not correlated with response status (Odds
blue in all groups. We found the largest relative change ratio 1.03, 95% CI 0.94–1.13; p = 0.4568).
from VIS 0 h to VIS 3 h (1VIS) in the patient cohort Next, we investigated whether clinical variables assessed at
treated with a methylene blue bolus only (1VIS 29 62%); the bedside are associated with a response to methylene blue. As
however, this did not differ significantly from the other shown in Table 3, cardiogenic shock and norepinephrine were
cohorts (bolus + continuous infusion 1VIS 29 ± 33%; associated with a higher chance of responding to methylene
continuous infusion only 11 ± 37%; one-way ANOVA blue administration, while the bolus treatment regimen was not
p = 0.1184). significantly associated with response.

Frontiers in Medicine 05 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

TABLE 2 ICU variables.

Total Bolus + Bolus only Continuous p values

(n = 209) infusion (n = 59) infusion only
(n = 111) (n = 39)

28-day mortality n (%) 130 (62.2) 59 (53.1) 42 (71.2) 29 (74.3) 0.0154

SAPS II (mean ± SD) 59.5 ± 12.0 59.5 ± 12.5 60.4 ± 12.6 61.9 ± 9.4 0.2091
APACHE II (mean ± SD) 25.8 ± 6.4 24.8 ± 6.1 26.9 ± 7.4 27.0 ± 5.4 0.0529
SOFA (mean ± SD) 11.3 ± 3.0 11.04 ± 3.1 11.3 ± 3.3 12.0 ± 2.5 0.2313
Renal replacement therapy in the ICU n (%) 183 (87.5) 98 (88.2) 50 (84.7) 35 (89.7) 0.7212
Median LOS-ICU of survivors (interquartile 22 (10–43) 11 (4–25) 7 (1–16) 4 (2–14) 0.0067
range)
Median hours of ventilator support (interquartile 114 (37–283) 160 (50–334) 113 (24–279) 66 (24–220) 0.0304#
range)
Cumulative dose of methylene blue
Methylen blue (mg/kg) 5.04 ± 3.2 6.7 ± 3.2 2.7 ± 1.9 4.03 ± 2.5 <0.0001
Initial doses of vasoactive substances
at methylene blue administration
VIS (interquartile range) [µg/kg/min] 59.8 (39.2–82.4) 58.0 (37.3–80.7) 60.3 (39.8–91.4) 59.6 (41.8–80.5) 0.4243
Norepinephrine treatment n (%) 209 (100) 111 (100) 59 (100) 39 (100) 1.0000
Norepinephrine dose (interquartile range) 0.47 (0.30–0.73) 0.46 (0.28–0.71) 0.53 (0.32–0.77) 0.45 (0.32–0.68) 0.2662
[µg/kg/min]
Epinephrine treatment n (%) 56 (27) 32 (29) 13 (22) 11 (28) 0.6202
Epinephrine dose (interquartile range) 0.043 (0.02–0.10) 0.037 (0.0–0.07) 0.088 (0.02–0.11) 0.10 (0.033–0.129) 0.8809
[µg/kg/min]
Vasopressin treatment n (%) 191 (91) 105 (94) 49 (83) 37 (94) 0.0265
Vasopressin dose (interquartile range) [U/kg/min] 3 (2–4) 3 (2–4) 3 (2–3) 3 (2–4) 0.5437
Dobutamine treatment n (%) 40 (19) 21 (18) 9 (15) 10 (25) 0.4395
Dobutamine dose (interquartile range) 5.0 (3.1–6.21) 4.9 (2.8–6.0) 5.59 (2.9–8.2) 5.0 (4.0–6.2) 0.9533
[µg/kg/min]
Milrinon treatment n (%) 118 (56) 66 (59) 32 (54) 20 (51) 0.6218
Milrinon dose (interquartile range) [µg/kg/min] 0.44 (0.3–0.5) 0.41 (0.3–0.5) 0.42 (0.3–0.5) 0.5 (0.4–0.6) 0.4052
Point of care diagnostics
Lactate (interquartile range) [mmol/L] 5.9 (3.3–11.2) 5.5 (3.1–9.8) 6.1 (4–10.7) 9.1 (3.5–12.9) 0.7743
pH (interquartile range) 7.32 (7.27–7.38) 7.33 (7.27–7.38) 7.31 (7.27–7.37) 7.32 (7.27–7.39) 0.6371

# Compared using Wilcoxon rank sum test.

In the multivariate model, cardiogenic shock, univariate analysis, we found that SAPSII, lactate at the start
norepinephrine dose at the time of methylene blue of methylene blue administration, blood pH, and higher VIS
administration, and treatment with bolus alone (negative (translated as higher doses of norepinephrine and vasopressin
association) remained independently associated with response doses) were associated with mortality at 28 days. Interestingly,
to methylene blue (Table 3). ROC analysis displayed a the administration of methylene blue as bolus + continuous
discrimination with an AUC of 0.68551. The goodness-of-fit infusion was associated with decreased mortality at 28 days. In a
was appropriate (Hosmer-Lemeshow; p = 0.8851). multivariate model, SAPSII, lactate, and methylene blue dosing
strategies remained independently associated with mortality
28 days after methylene blue administration (AUC 0.8054).
Variables associated with mortality at 28 The Hosmer-Lemeshow test showed appropriate goodness-of-fit
days (p = 0.8180).
In summary, regarding the methylene blue treatment
The association of variables with mortality at 28 days in regimen, only bolus administration with continuous infusion
the logistic regression analysis is shown in Table 4. In the was associated with a reduced mortality at 28 days.

Frontiers in Medicine 06 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

FIGURE 2
Hemodynamic changes after methylene blue administration. The injection time points of the methylene blue bolus were defined as “0 h”; all
subsequent time indicators were defined in relation to the administration of methylene blue administration. (A) Mean arterial pressure over time
in the different methylene blue treatment groups over time. (B) Heart rate over time after methylene blue administration (A + B: bolus +
continuous infusion: n = 111–110; bolus only: n = 47–59; continuous infusion only n = 35–39). (C) Norepinephrine infusion rate in µg/kg/min
over time after methylene blue administration. (D) Development of vasoactive inotropic score after methylene blue administration (C + D: bolus
+ continuous infusion: n = 111–110; bolus only: n = 47–59; continuous infusion only n = 35–39); All data are shown as mean ± 95%
confidence interval.

Discussion Shock remains a significant problem in critically ill

patients, with high inherent mortality. An international
In this retrospective cohort study, we found that methylene consensus conference defined shock as a “life-threatening
blue reduces the requirement for vasoactive substances, generalized maldistribution of blood flow that results in
regardless of the dosing strategy. When analyzing the factors the inability to deliver and/or use an adequate amount of
associated with the response to methylene blue, we found that oxygen, leading to tissue dysoxia” (16). Numerous basic
the response rate to methylene blue increased in patients with science and clinical studies investigated the underlying
cardiogenic shock and with higher doses of norepinephrine. dysregulation of endogenous pathways that sustain defective
Administration of a methylene blue bolus without subsequent hemodynamics in shock. Nitric oxide has gained substantial
continuous infusion reduced the response to methylene blue. research interest because it influences multiple pathways
The administration of methylene blue as a bolus followed by associated with the development of shock. In homeostasis,
continuous infusion decreased 28-day mortality. Importantly, NO regulates blood flow in resistance vessels (17), influences
no serious undesired effects were observed in our cohort. We coagulation, and modifies inflammatory pathways. During
only noticed minor undesired drug effects such as skin or shock, NO production increases dramatically up to
urine discoloration. 1000 times (1, 18).

Frontiers in Medicine 07 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

and plasma levels of the pro-inflammatory cytokine IL-6 in an

animal with septic shock (23).
In our cohort, the ICU scoring systems on the day of
methylene blue treatment implied a mortality of 50–70% in all
groups based on the mean of SAPSII, APACHE III and SOFA.
Interestingly, the measured mortality at 28 days in patients
receiving a bolus + continuous infusion was at the lower end of
the predicted mortality compared to the other dosing regimens.
In a univariate and multivariate logistic regression analysis,
bolus administration followed by continuous infusion was
associated with a lower mortality at 28 days. Interestingly, the
cumulative dose of methylene blue, did not differ between the
different treatment regimen groups. This could suggest that not
the amount of methylene blue per se, but also pharmacokinetic
properties could influence the response to methylene blue. As
such, this observation is well known in critically ill patients
in an entirely different context, e.g., treatment with beta-
lactam anti-infective drugs. Here, the clinical response rate also
increases when substances are administered with a continuous
infusion. The cumulative dose does not differ between the
different forms of administration. Therefore, the observed effect
that methylene blue response increases with a bolus injection
(corresponding to a loading dose in anti-infective therapy)
and subsequent continuous infusion, could point into the
FIGURE 3
Percentage of methylene blue responders per administration direction of a pharmacodynamic effect that influence the effect
group. (A) Schematic of the definition of responder or of methylene blue. However, prospective studies assessing this
non-responder status to methylene blue treatment. (B)
phenomenon are lacking.
Proportion of responders and non-responders in the study
cohort (n = 196 in total) (C) bolus + continuous infusion cohort Interestingly, the reduction in mortality in our cohort
(n = 110 in total) (D) bolus only infusion group (n = 48) or (E) appeared to be not related to vasopressor requirements
continuous only cohort (n = 38).
because the three groups experienced similar hemodynamic
stabilization. Methylene blue as a bolus and subsequent
continuous infusion could cause a sustained inhibition of the
NO pathway, as methylene blue blocks the NO pathway in two
In 1992, a case series reported that hemodynamics stages: it inhibits both NO synthase (24) and the downstream
improved in patients with shock after patients received the effector enzyme, soluble guanylate cyclase (sGC) (25). However,
NO pathway inhibitor methylene blue (19). Since then, several methylene blue also interferes with numerous other pathways
prospective randomized controlled trials have investigated (26), and therefore the beneficial properties observed could
whether methylene blue reduces vasopressor requirements (11, induce effects in addition to hemodynamics or NO-path
12, 19, 20). One trial reported that in patients with vasoplegia inhibition alone. In fact, inhibition of NO-synthase leaves shock
after cardiac surgery, mortality in the methylene blue treated mortality unchanged (27). The methylene blue concentration in
group decreased to 0 vs. 21.4% in the placebo treated group (12). whole blood exceeds the plasma concentration, indicating that
However, the study included only 28 patients per group, limiting the substance accumulates in blood cells. In acute inflammation,
the generalizability of the trial. immune cells, such as macrophages, release NO (25), thus
Based on the mechanism of proposed effect in hemodynamic enhancing the inflammatory response (28). The accumulation
instability (inhibition of NO-production), it is surprising of methylene blue in these cells could dampen an overshooting
that success of methylene blue treatment was not superior inflammatory response. In particular, in septic shock, NO
in vasoplegic shock. This suggests that methylene could overproduction contributes to hypotension, cardiodepression,
elicit beneficial effect different from NO-inhibition. One and vascular hyporeactivity (29). Therefore, the observed effects
potential explanation could be that methylene blue acts as an of methylene blue could modulate inflammatory responses and
antioxidant, similar to vitamin C. Shock patients suffer from subsequently reduce mortality.
an overall depletion of antioxidants (21), and administration In our analysis, the response to methylene blue
of antioxidants have been suggested as a therapeutic strategy administration was correlated with the diagnosis of cardiogenic
(22). In fact, antioxidants reduce the oxidative stress parameter, shock. This seems counterintuitive at first because, due to

Frontiers in Medicine 08 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

TABLE 3 Factors associated with methylene blue response in the logistic regression.

Univariate logistic regression Multivariate logistic regression

Odds ratio (95%CI) p-value Odds ratio (95%CI) p-value

ICU variables on day of methylene blue treatment

Age 0.97 (0.22–4.26) 0.9686
Sex 1.65 (0.84–3.26) 0.1378
SAPSII 0.99 (0.97–1.01) 0.7118
Septic shock 0.67 (0.37–1.12) 0.1903
Cardiogenic shock 2.10 (1.09–4.05) 0.0230 2.21 (1.11–4.36) 0.0224
Vasoplegia of other causes 0.77 (0.41–1.41) 0.4010
Point-of-care 0h after methylene blue treatment
Lactate 1.00 (0.95 to 1.06) 0.7722
pH 0.38 (0.009 to 15.27) 0.6122
Vasopressors/inotropes 0 h after methylene blue treatment
Vasoactive Inotropic Score 1.01 (1.00–1.02) 0.029
Norepinephrine µg/kg/min) 3.75 (1.46–9.62) 0.0023 3.64 (1.39–9.51) 0.0083
Dobutamin µg/kg/min)e 1.00 (0.95–1.05) 0.0317
Vasopressin µg/kg/min) 1.21 (0.95–1.54) 0.0938
Milrinon µg/kg/min) 0.17 (1.28–5.77) 0.0794
Hemodynamic variables
Mean arterial pressure 1.00 (0.97–1.02) 0.8052
Heart rate 0.99 (0.98–1.01) 0.8175
Heart index 1.22 (0.88–1.74) 0.1939
Systemic Vascular Resistance 0.99 (0.99–1.00) 0.2189
Dosing strategy
Bolus + continuous infusion 1.52 (0.85–2.7) 0.1517
Bolus only 0.48 (0.23–0.98) 0.0452 0.46 (0.22–0.99) 0.0483
Continuous infusion only 1.07 (0.55–2.08) 0.8413

Italic letters indicate statistical significance.

the mechanism of the proposed action, most of the studies performed our study under the assumption that methylene blue
investigated methylene blue in vasoplegia. However, there decreases the vasopressor requirements. We have not included a
is some evidence that increased NO release enhances the control group with shock and without methylene blue treatment,
detrimental spiral of events in cardiogenic shock. As such, thus our study cannot answer cannot provide evidence if
NO blocks the adrenergic response of the myocardium methylene reduces mortality per se. However, our findings
and reduces myocardial contractility (30). Similarly, NO suggest that within differences could exist between different
concentration increases in the peripheral blood of patients with administration protocols. Being retrospective in design, our
acute myocardial ischemia (31). Furthermore, 20% of patients study has the inherent limitation that we cannot account for all
with underlying cardiogenic shock experience a systemic variables that impacted the prescription and timing of methylene
inflammation response syndrome, and the majority show a blue administration. As such, blood transfusion or volume
positive blood culture with low systemic vascular resistance (32). resuscitation could have changed the findings of our study.
This provides an explanation why patients with the primary Nevertheless, disease severity and forms of shock were similar
diagnosis of cardiogenic shock could benefit from methylene between all groups, which suggests that differences between
blue treatment. different treatment groups were comparable. In addition, we
did not assess the length of stay in the ICU prior to methylene
blue treatment. Another important limitation is the size of
Limitations the individual cohorts. Even though we did not detect a
significant difference in demographic parameters or in ICU
Finally, our study has several important limitations. Our scoring systems, the numeric differences between the groups are
results should be seen in the context of the study design. First, we notable. Moreover, we did not perform retrospective matching

Frontiers in Medicine 09 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

TABLE 4 Factors associated with mortality at 28 days in logistic regression.

Univariate logistic regression Multivariate logistic regression

Odds ratio (95%CI) p-value Odds ratio (95%CI) p-value

ICU variables on day of methylene blue treatment

Age 2.91 (0.69–12.3) 0.1450
Sex 1.45 (0.75–2.80) 0.2622
SAPSII 1.07 (1.04–1.10) <0.001 1.08 (1.04–1.12) <0.0001
Septic shock 1.54 (0.86–2.76) 0.1424
Cardiogenic shock 1.12 (0.61–2.06) 0.7118
Vasoplegia of other causes 0.55 (0.30–1.01) 0.0561
Point-of-care 0h after methylene blue treatment
Lactate 1.18 (1.09–1.26) <0.001 1.16 (1.08–1.25) <0.0001
pH 0.0014 (3,123E-5–0.071 0.0006 0.08 (0.0006–10.05) 0.3054
Vasopressors/inotropes 0 h after methylene blue treatment
Vasoactive Inotropic Score 1.01 (1.00–1.02) 0.0004
Norepineprhine (µg/kg/min) 4.56 (1.75–11.88) 0.0004 3.06 (0.97–9.64) 0.0559
Dobutamine (µg/kg/min) 1.03 (0.88–1.20) 0.6778
Vasopressin (µg/kg/min) 1.28 (1.00–1.63) 0.0314
Milrinon(µg/kg/min) 1.12 (0.15–8.07) 0.9102
Hemodynamic variables 0 h after methylene blue treatment
Mean arterial pressure 0.99 (0.97–1.02) 0.9077
Heart rate 1.00 (0.98–1.01) 0.6100
Heart index 0.97 (0.71–1.32) 0.8704
Systemic vascular resistance 1.00 (0.99–1.00) 0.6818
Dosing strategy of methylene blue
Bolus + continuous infusion 0.43 (0.24–0.76) 0.0039 0.45 (0.23–0.90) 0.0246
Bolus only 1.98 (0.90–4.32) 0.0759
Continuous infusion only 1.74 (0.90–3.33) 0.8889

Italic letters indicate statistical significance.

(such as propensity score matching) as this technique does not Data availability statement
account for unmeasured characteristics and cofounders (33).
Despite these limitations, our study adds to the body of The original contributions presented in the study are
evidence on the effect of methylene blue on shock. To our included in the article/supplementary material, further inquiries
knowledge, our study presents the largest cohort of methylene can be directed to the corresponding author.
blue treatment in ICU patients to date. Thus, we have concluded
that methylene blue is safe in this population and increases
survival in some patients. Ethics statement
The approval for this study was obtained from the local
Conclusion institutional review board and the Ethics Committee of the
Tübingen University Hospital Tübingen (549/2019BO2).
In summary, we report the results of the largest
cohort of methylene blue-treated ICU patients to date. In
our single-center, retrospective analysis, methylene blue Author contributions
administration was associated with a reduced 28-day mortality
in critically ill patients with shock, when administered SS-Y, PW, and MKo designed the study. MKo performed
as a bolus followed by a continuous infusion. However, data analysis and interpretation, and statistical data analysis.
prospective randomized studies are needed to systematically HM and MKe performed statistical analysis and helped to write
evaluated the values of methylene blue in the treatment the manuscript. Y-SF cross-checked statistical analysis. K-LH-S,
of shock. CS, and PR contributed to the manuscript preparation, drafting,

Frontiers in Medicine 10 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

critique, and review. All authors approved the final version of that could be construed as a potential conflict
the manuscript submitted. of interest.

Acknowledgments
Publisher’s note
This work was supported by the Deutsche
All claims expressed in this article are solely those
Forschungsgemeinschaft grant (KO3884/5-1).
of the authors and do not necessarily represent those
of their affiliated organizations, or those of the publisher,
Conflict of interest the editors and the reviewers. Any product that may be
evaluated in this article, or claim that may be made by
The authors declare that the research was conducted in its manufacturer, is not guaranteed or endorsed by the
the absence of any commercial or financial relationships publisher.

References
1. Kimmoun A, Levy B. Pathophysiology of Shock Oxford 15. Belletti A, Lerose CC, Zangrillo A, Landoni G. Vasoactive-inotropic score:
Textbook of Critical Care. Oxford, UK: Oxford University Press evolution, clinical utility, and pitfalls. J Cardiothorac Vasc Anesth. (2021) 35:3067–
(2016). doi: 10.1093/med/9780199600830.003.0149 77. doi: 10.1053/j.jvca.2020.09.117
2. Cecconi M, De Backer D, Antonelli M, Beale R, Bakker J, Hofer C, et al. 16. Antonelli M, Levy M, Andrews PJ, Chastre J, Hudson LD. International
Consensus on circulatory shock and hemodynamic monitoring. Task force of the consensus conference p. France, 27–28 April 2006. Intensive Care Med. (2007)
European society of intensive care medicine. Intens Care Med. (2014) 40:1795– 33:575–90. doi: 10.1007/s00134-007-0531-4
815. doi: 10.1007/s00134-014-3525-z
17. Koeppen M, Gravlee GP, Nasrallah F, Eckle T. Transesophageal
3. Koeppen M, Feil R, Siegl D, Feil S, Hofmann F, Pohl U, et al. echocardiography in the diagnosis of acute pericardial tamponade
cGMP-dependent protein kinase mediates NO- but not acetylcholine- during hiatal hernia repair. J Cardiothorac Vasc Anesth. (2014)
induced dilations in resistance vessels in vivo. Hypertension. (2004) 28:112–4. doi: 10.1053/j.jvca.2012.09.015
44:952–5. doi: 10.1161/01.HYP.0000147661.80059.ca
18. Wildhirt SM, Dudek RR, Suzuki H, Bing RJ. Involvement of inducible nitric
4. Joffre J, Hellman J, Ince C, Ait-Oufella H. Endothelial responses in sepsis. Am oxide synthase in the inflammatory process of myocardial infarction. Int J Cardiol.
J Respir Crit Care Med. (2020) 202:361–70. doi: 10.1164/rccm.201910-1911TR (1995) 50:253–61. doi: 10.1016/0167-5273(95)02385-A
5. Petros A, Lamb G, Leone A, Moncada S, Bennett D, Vallance P. Effects of a 19. Schneider F, Lutun P, Hasselmann M, Stoclet JC, Tempé JD. Methylene
nitric oxide synthase inhibitor in humans with septic shock. Cardiovasc Res. (1994) blue increases systemic vascular resistance in human septic shock. Preliminary
28:34–9. doi: 10.1093/cvr/28.1.34 observations. Intensive Care Med. (1992) 18:309–11. doi: 10.1007/BF01706481
6. Myburgh JA, Higgins A, Jovanovska A, Lipman J, Ramakrishnan N, 20. Preiser JC, Lejeune P, Roman A, Carlier E, De Backer D, Leeman M, et al.
Santamaria J. A comparison of epinephrine and norepinephrine in critically ill Methylene blue administration in septic shock: a clinical trial. Crit Care Med.
patients. Intens Care Med. (2008) 34:2226–34. doi: 10.1007/s00134-008-1219-0 (1995) 23:259–64. doi: 10.1097/00003246-199502000-00010
7. Annane D, Vignon P, Renault A, Bollaert PE, Charpentier C, 21. Goode HF, Cowley HC, Walker BE, Howdle PD, Webster NR. Decreased
Martin C, et al. Norepinephrine plus dobutamine vs. epinephrine alone antioxidant status and increased lipid peroxidation in patients with septic
for management of septic shock: a randomised trial. Lancet. (2007) shock and secondary organ dysfunction. Crit Care Med. (1995) 23:646–
370:676–84. doi: 10.1016/S0140-6736(07)61344-0 51. doi: 10.1097/00003246-199504000-00011
8. De Backer D, Biston P, Devriendt J, Madl C, Chochrad D, Aldecoa C, et al. 22. Pennington CR. Disease-associated malnutrition in the year 2000. Postgrad
Comparison of dopamine and norepinephrine in the treatment of shock. N Engl J Med J. (1998) 74:65–71. doi: 10.1136/pgmj.74.868.65
Med. (2010) 362:779–89. doi: 10.1056/NEJMoa0907118
23. Lowes DA, Webster NR, Murphy MP, Galley HF. Antioxidants that protect
9. Bailey A, Pope TW, Moore SA, Campbell CL. The Tragedy of TRIUMPH for mitochondria reduce interleukin-6 and oxidative stress, improve mitochondrial
nitric oxide synthesis inhibition in cardiogenic shock. Am J Cardiovascular Drugs. function, and reduce biochemical markers of organ dysfunction in a rat model of
(2007) 7:337–45. doi: 10.2165/00129784-200707050-00003 acute sepsis. Br J Anaesth. (2013) 110:472–80. doi: 10.1093/bja/aes577
10. Howland RH. Methylene blue: the long and winding road from 24. Mayer B, Brunner F, Schmidt K. Inhibition of nitric oxide
stain to brain: part 1. J Psychosoc Nurs Ment Health Serv. (2016) 54:21– synthesis by methylene blue. Biochem Pharmacol. (1993) 45:367–
4. doi: 10.3928/02793695-20160818-01 74. doi: 10.1016/0006-2952(93)90072-5
11. Kirov MY, Evgenov OV, Evgenov NV, Egorina EM, Sovershaev MA, 25. Martin W, Villani GM, Jothianandan D, Furchgott RF. Selective blockade of
Sveinbjørnsson B, et al. Infusion of methylene blue in human septic shock: endothelium-dependent and glyceryl trinitrate-induced relaxation by hemoglobin
a pilot, randomized, controlled study. Crit Care Med. (2001) 29:1860– and by methylene blue in the rabbit aorta. J Pharmacol Exp Ther. (1985) 232:708–
7. doi: 10.1097/00003246-200110000-00002 16.
12. Levin RL, Degrange MA, Bruno GF, Del Mazo CD, Taborda 26. Oz M, Lorke DE, Hasan M, Petroianu GA. Cellular and molecular
DJ, Griotti JJ, et al. Methylene blue reduces mortality and morbidity actions of methylene blue in the nervous system. Med Res Rev. (2011) 31:93–
in vasoplegic patients after cardiac surgery. Ann Thorac Surg. (2004) 117. doi: 10.1002/med.20177
77:496–9. doi: 10.1016/S0003-4975(03)01510-8
27. Alexander JH, Reynolds HR, Stebbins AL, Dzavik V, Harrington RA,
13. Hohlfelder B, Douglas A, Wang L, Wanek M, Bauer SR. Van de Werf F, et al. Effect of tilarginine acetate in patients with acute
Association of methylene blue dosing with hemodynamic response for the myocardial infarction and cardiogenic shock: the TRIUMPH randomized
treatment of vasoplegia. J Cardiothorac Vasc Anesth. (2022) 36:3543–50. controlled trial. JAMA. (2007) 297:1657–66. doi: 10.1001/jama.297.15.joc
doi: 10.1053/j.jvca.2022.04.003 70035
14. Porizka M, Kopecky P, Dvorakova H, Kunstyr J, Lips M, Michalek P, et al. 28. Moilanen E, Vapaatalo H. Nitric oxide in inflammation and
Methylene blue administration in patients with refractory distributive shock—a immune response. Ann Med. (1995) 27:359–67. doi: 10.3109/078538995090
retrospective study. Sci Rep. (2020) 10:1828. doi: 10.1038/s41598-020-58828-4 02589

Frontiers in Medicine 11 frontiersin.org

Sari-Yavuz et al. 10.3389/fmed.2022.1014276

29. Kirkebøen KA, Strand OA. The role of nitric oxide in heart. J Am Coll Cardiol. (1998) 32:373–9. doi: 10.1016/S0735-1097(98)0
sepsis—an overview. Acta Anaesthesiol Scand. (1999) 43:275– 0270-8
88. doi: 10.1034/j.1399-6576.1999.430307.x
32. Kohsaka S, Menon V, Lowe AM, Lange M, Dzavik V, Sleeper LA,
30. Keaney JF, Hare JM, Balligand JL, Loscalzo J, Smith TW, Colucci et al. Systemic inflammatory response syndrome after acute myocardial
WS. Inhibition of nitric oxide synthase augments myocardial contractile infarction complicated by cardiogenic shock. Arch Intern Med. (2005) 165:1643–
responses to beta-adrenergic stimulation. Am J Physiol. (1996) 271:H2646– 50. doi: 10.1001/archinte.165.14.1643
2652. doi: 10.1152/ajpheart.1996.271.6.H2646
33. Nuttall Gregory A, Houle Timothy T. Liars, damn liars, and propensity
31. Akiyama K, Kimura A, Suzuki H, Takeyama Y, Gluckman TL, Terhakopian scores. Anesthesiology. (2008) 108:3–4. doi: 10.1097/01.anes.0000296718.35
A, et al. Production of oxidative products ofnitric oxide in infarcted human 703.20

Frontiers in Medicine 12 frontiersin.org