Article, Emergency Medicine

Beta-blockers’ effect on Levels of Lactate in patients with suspected sepsis – The BeLLa study

a b s t r a c t

Objective: In the assessment and management of septic patients in the emergency department (ED), serum lac- tate is often measured to stratify severity to guide decision making. Increased adrenergic drive has been postu- lated as a contributory factor for hyperlactatemia in sepsis. We aim to prospectively evaluate the effect of chronic beta-blocker use on Serum lactate levels in sepsis at initial presentation to the ED. Methods: We conducted a prospective observational study at the ED of a tertiary care academic medical center in Singapore. One hundred and ninety-five ED patients who fulfilled all of the following: (1) age 45 years and above,

(2) tympanic temperature >= 37.8 ?C or clinically suspected to have an infection, and (3) quick Sequential (Sepsis- Related) Organ Failure Assessment (qSOFA) score >= 1 were included in the study. Serum venous lactate was sam- pled within two hours from presentation to the ED. The primary outcome measure was the difference in initial serum venous lactate concentration at presentation to the ED in patients on chronic beta-blockers versus patients without.

Results: Seventy patients (35.9%) were on long-term beta-blocker therapy. The primary outcome of mean initial serum Venous lactate concentration was similar between patients prescribed chronic beta-blocker therapy and patients without (1.78 mmol/L versus 1.70 mmol/L, p = .540). Chronic beta-blocker therapy also did not signif- icantly affect mean initial serum venous lactate concentration across all subgroups of sepsis risk stratification. Conclusions: Long-term beta-blocker therapy did not significantly affect initial serum venous lactate concentra- tion in ED patients with suspected sepsis.

(C) 2019

  1. Introduction

Abbreviations: ED, emergency department; qSOFA, quick Sequential (Sepsis-Related) Organ Failure Assessment.

* Corresponding author at: Emergency Medicine Department, National University Hospital, 9 Lower Kent Ridge Road, Level 4, Singapore 119085, Singapore.

E-mail address: [email protected] (M.T. Chua).

1 Present address: Woodlands Health Campus, 17 Woodlands Drive, Singapore 738097.

2 Previous address: Emergency Medicine Department, National University Hospital, 9 Lower Kent Ridge Road, Level 4, Singapore 119085.

3 1st author’s old affiliation: Emergency Medicine Department, National University Hospital, National University Health System, Singapore.

4 Duke-NUS Medical School, 8 College Rd, Singapore 169857.

5 Emergency Medicine Department, National University Hospital, 9 Lower Kent Ridge Road, Level 4, Singapore 119085.

6 Ministry of Health Holdings Pte Ltd, 1 Maritime Square, Singapore 099253.

Sepsis occurs when an infection is complicated by life-threatening organ dysfunction secondary to maladaptive host responses [1]. It is commonly encountered in the emergency department (ED) and is asso- ciated with significant mortality [2]. Early recognition and treatment of sepsis is crucial to improve patient outcomes. Serum lactate is often measured and used in conjunction with the patient’s clinical state for risk stratification to prognosticate sepsis and guide decision making [3]. It was traditionally believed that hyperlactatemia in sepsis is mainly driven by reduced tissue oxygen delivery secondary to hypoperfusion, resulting in anaerobic lactate production [4]. However, serum lactate el- evation may not necessarily indicate tissue hypoxia [5]. Several alterna- tive explanations have since been proposed, including beta-adrenergic stimulation [4], microcirculatory dysfunction [6] and reduced Lactate elimination [7]. Evidence supporting beta-adrenergic stimulation in

https://doi.org/10.1016/j.ajem.2019.12.046

0735-6757/(C) 2019

septic shock include increased serum lactate levels following epineph- rine infusion [8] and decreased levels after esmolol infusion despite re- duction in tissue oxygen delivery [9,10]. It has been postulated that increased adrenergic drive during sepsis could contribute to hyperlactatemia under aerobic conditions via beta-2 adrenergic recep- tor activation with increased downstream glycolytic flux [5,11].

A retrospective study by Contenti and colleagues reported that sep-

tic patients on chronic beta-blockers had lower mean serum lactate when compared to those who were not [12]. However, use of beta ago- nist prior to serum lactate sampling and compliance to long-term beta- blocker therapy were not considered. In another retrospective study, septic patients on long-term beta-blockers had lower 28-day mortality as compared to patients without, despite having higher risk profiles [13]. However, this study only included intensive care unit (ICU) pa- tients and could be susceptible to selection bias.

Therefore, a prospective study would be useful to address the limita- tions of aforementioned studies and to evaluate any association be- tween chronic beta-blocker use and serum lactate concentrations in septic patients. Any such association should then be taken into consid- eration by ED physicians for diagnosis, risk stratification and subsequent management for septic patients on chronic beta-blocker therapy.

The objective of this study was to assess the effect of chronic beta- blocker use on serum venous lactate in suspectED septic patients at ED presentation. We hypothesize that chronic beta-blocker use would re- duce serum venous lactate in potentially septic ED patients by

1.5 mmol/L or more compared to patients without chronic beta- blocker therapy. Secondary objectives were to ascertain the differences in ICU admission rates and 28-day all-cause mortality in patients on chronic beta-blocker therapy compared to those without.

  1. Methods
    1. Study design and setting

This prospective observational study was conducted at the ED of Na- tional University Hospital, a 1200-bed tertiary academic medical center in Singapore, between March 2017 and August 2018. Our ED receives over 110,000 visits annually, of which 47% of the cases require urgent (42.5%) or immediate (4.5%) care. Informed consent was obtained from patients or their legally acceptable representative (LAR) if patients lacked mental capacity at presentation. For patients who lacked mental capacity with no accompanying LAR, declaration for waiver of initial consent at time of enrolment was sought from an independent ED spe- cialist not part of the study team. Delayed informed consent was later obtained from the patient after regaining mental capacity or their LAR. The beta-blockers’ effect on levels of lactate (BeLLa) study protocol was approved by the local institutional ethics review board.

Selection of participants

Patients presenting to the ED were screened at triage, from 0800 h to 1800 h on weekdays and occasionally on weekends when a co- investigator was on-site. Those who fulfilled all of the following criteria were invited to participate: (1) age >= 45 years, (2) tympanic temperature >= 37.8 ?C or clinically suspected to have an infection and

(3) quick Sequential (Sepsis-Related) Organ Failure Assessment (qSOFA) score >= 1 (comprising respiratory rate >= 22, altered mentation [Glasgow Coma Score b 15] and systolic blood pressure <= 100 mmHg) [1]. This age group was selected because patients aged >=45 years were more likely to have chronic medical conditions requiring long-term pharmacotherapy [14,15].

Patients with any of the following were excluded from the study:

(1) Child-Turcotte-Pugh class B and C Chronic liver disease, (2) scleral icterus at presentation, (3) on metformin therapy, (4) received beta- adrenergic agonist nebulization 4 h prior to blood sampling, (5) refusal of consent and (6) do-not-resuscitate status.

Sample size calculation

Using data from a previous randomized controlled trial on patients with sepsis from the same center [16], we estimated the prevalence of chronic beta-blocker therapy to be 25%. Based on previous related stud- ies by Contenti et al. [12] and Fuchs et al. [17], we estimated a mean serum venous lactate difference of 1.5 mmol/L. For this study to achieve 80% power and alpha of 5%, assuming a mean difference of 1.5 mmol/L and a standard deviation of 3.0 mmol/L, a total sample size of at least 172 patients (with and without beta-blocker therapy) was required.

Measurements

Serum venous lactate was sampled within two hours from ED pre- sentation, along with other blood investigations deemed necessary by the attending physician as per standard of care.

Variables collected

A standardized case report form was utilized to obtain demo- graphics including age, gender, race, nursing home residence, past med- ical history, vital signs, Glasgow Coma Score at triage, results of ED investigations including initial serum venous lactate, source of infec- tion based on final diagnosis at discharge, ICU admission and all-cause 28-day mortality from the hospital’s electronic medical records. Details of chronic beta-blocker therapy were retrieved from electronic medical records and verified with patients or their LAR. Information about com- pliance to chronic beta-blocker therapy, including the date and time of the last dose of beta-blocker before presentation, were obtained from patients or their LAR where possible.

Risk stratification of sepsis was performed using three scoring sys- tEMS used to predict mortality in clinical settings – SOFA score [1,18], Mortality in Emergency Department Sepsis (MEDS) score [19,20] and Predisposition, Insult and Response in Organ Failure (PIRO) score [21]. A modified PIRO score that excludes two variables, heart rate and serum lactate, was also calculated, as these variables could be affected by chronic beta-blocker therapy. Patients were then stratified into sub-groups based on mortality as reflected by their SOFA [22], MEDS

[20] and PIRO [21] scores. Normal bilirubin concentrations were as- sumed for patients without documented clinical jaundice and had left the ED without requiring such a test as determined by the attending physician.

For patients with multiple sources of infection based on the final di- agnosis at discharge, the primary source of infection was determined after adjudication by two independent emergency medicine specialists (MTC and WSK). A third independent specialist (KSL) was consulted when opinions of the first two adjudicators differed.

Outcomes

The primary outcome measure was the difference in initial serum venous lactate at presentation to the ED in patients on chronic beta- blockers versus patients without. Secondary outcome measures were 28-day all-cause mortality and ICU admission rates.

Statistical analysis

Data analysis was performed using STATA 15 (Stata Corp LP, College Station, TX, USA). Parametric data are reported as means with standard deviations (SD) and analyzed using the Student’s t-test. Non-parametric data are reported as medians with interquartile ranges (IQR) and ana- lyzed with the Mann-Whitney U test. Categorical data are reported as proportions with percentages and analyzed with chi-square or Fisher’s exact test, as appropriate. Multivariate linear regression analysis was performed to determine independent variables associated with initial serum venous lactate concentration. A forward stepwise approach was

used, including only variables with p-value b.10 into the model. A p- value of b0.05 was considered statistically significant.

Table 1

Mean initial serum venous lactate concentration difference.

Type of analysis Beta-blockers No beta-blockers P

Sensitivity analyses

N (%) Lactate, mean (SD)

N (%) Lactate, mean (SD)

value

Two sensitivity analyses were planned post-hoc to take into account compliance and the last dose of beta-blockers prior to ED presentation.

Main analysis N = 70 N = 125 0.540

70 (35.9) 1.70 (1.54) 125 (64.1) 1.78 (1.71)

Description of comparator groups for the primary analysis and the two sensitivity analyses are detailed in Table 1 of Supplementary Material. In brief, the first sensitivity analysis was designed to determine the effect

Sensitivity analysis 1a

Sensitivity analysis 2a

N = 57 N = 138 0.110

57 (29.2) 1.61 (1.48) 138 (70.8) 1.82 (1.70)

N = 58 N = 131 0.202

58 (30.7) 1.62 (1.50) 131 (69.3) 1.79 (1.69)

of regular long-term beta-blocker therapy, with compliance to therapy

defined as not missing more than one dose per week. The second sensi- tivity analysis took pharmacokinetics of beta-blockers into consider- ation, with patients prescribed with a beta-blocker, regardless of compliance, classified as “no beta-blockers” if they presented to the ED beyond five half-lives from their last dose, as beta-blockers are ex- pected to be almost completely eliminated from the patient after five half-lives [23-26].

  1. Results
    1. Characteristics of study objects

Over the 18-month study period, 576 patients met the inclusion criteria. Among them, 358 patients (62.2%) met specific exclusion crite- rion (Fig. 1). Of the remaining 195 patients enrolled, 70 patients (35.9%) were on long-term beta-blocker therapy.

Patients on chronic beta-blocker therapy were older and had signif- icantly lower median initial heart rate (p b .001) (Table 2, Supplemen- tary Material). In both groups, respiratory infection was the most common source of infection (53.3%). The median SOFA score was higher in patients on chronic beta-blocker therapy (p b .001). Comorbidities such as hypertension, dyslipidemia, diabetes mellitus, ischemic heart disease and Renal impairment were also more common among patients on long-term beta-blockers, while malignancy was more common among patients without chronic beta-blocker therapy (Table 2, Supple- mentary Material).

Main results

The primary outcome of mean initial serum venous lactate concen- tration was similar between patients who were prescribed chronic beta-blocker therapy and patients without (1.78 mmol/L versus 1.70 mmol/L, p = .540) (Table 1). In the first sensitivity analysis, pa- tients who were prescribed and compliant to long-term beta-blocker therapy had similar mean initial lactate concentration compared to those prescribed but were non-compliant combined with patients who were not on long-term beta-blocker therapy (Table 1). In the sec- ond sensitivity analysis that takes the last dose of beta-blocker prior to presentation into account, mean initial lactate concentration was also similar between the two groups (Table 1).

Fig. 1. Flowchart illustrating patient screening and enrolment.

N – number of patients; SD – standard deviation.

a Description of comparator groups can be found in Table 1 of Supplementary Material.

Chronic beta-blocker therapy did not significantly affect mean initial lactate concentrations across all subgroups of sepsis risk stratification for mortality using the SOFA, MEDS and PIRO scores (Table 2). There was a trend towards rising lactate levels with increasing severity of sep- sis scores.

In the univariate analyses, ICU admission, 28-day all-cause mortality, heart rate, GCS, qSOFA and PIRO scores were significantly associated with initial lactate concentration (Table 3). After adjustment with mul- tivariate linear regression, only heart rate, ICU admission and PIRO score remained independently associated with initial lactate concentration (Table 3). Chronic beta-blocker therapy was not significantly associated with ICU admission rates and 28-day all-cause mortality (Table 4).

  1. Discussion

This study showed that mean serum venous lactate concentration at initial ED presentation was similar between patients prescribed with long-term beta-blocker therapy and patients without, even after adjusting for known confounders. Both sensitivity analyses were consis- tent with the primary analysis.

Hyperlactatemia in sepsis could arise due to an increased lactate production, a decreased lactate elimination and utilization, or a combi- nation of both [11]. Although there is no consensus on the cause of hyperlactatemia in sepsis, existing literature favors mechanisms caus- ing increased lactate production as the main driver of hyperlactatemia in early sepsis, with reduced Lactate clearance predominating in later stages [7]. Our study results suggest that mechanisms other than beta- adrenergic stimulation are likely to be involved in increased lactate pro- duction in sepsis.

Increased lactate production in sepsis was theorized to be due to an- aerobic respiration secondary to systemic hypoperfusion. Along this line of thought, the Surviving Sepsis Campaign advocates administration of intravenous fluids, vasopressor use and blood product transfusion for Circulatory support, with normalization of serum lactate viewed as a marker indicating resolution of tissue hypoxia [2]. This thought was challenged after serum lactate was found to remain elevated in septic patients despite achieving tissue oxygen delivery above the threshold for anaerobic respiration [5].

Subsequently, a concept termed microcirculatory and mitochondrial distress syndrome was proposed to account for lactate production de- spite sufficient tissue oxygen delivery [6]. In this syndrome, anaerobic lactate production might be triggered at the microcirculatory level (ar- terioles, capillaries and veins) due to mismatch in oxygen demand and supply secondary to inflammation-induced endothelial dysfunction against the backdrop of normal systemic oxygen delivery. Widespread tissue hypoxia is unlikely in our cohort considering the relatively high mean arterial pressure used as a surrogate measure of systemic perfu- sion, suggesting that a certain extent of microcirculatory dysfunction may be the cause of hyperlactatemia among our patients.

Piecing various hypotheses together, it appears that different mech- anisms might be at play depending on the stage of sepsis [11] – hyperlactatemia in early sepsis prior to resuscitation could largely be

Table 2

Mean initial serum venous lactate concentration after subgroup analyses by severity scores.

Severity scores

Beta-blockers

No beta-blockers

P value

N (% within subgroup)

Lactate, mean (SD)

N (% within subgroup)

Lactate, mean (SD)

SOFA (% of cohort)

0 to 6 (95.9)

66 (35.3)

1.68 (1.54)

121 (64.7)

1.77 (1.71)

0.478

7 to 9 (4.1)

4 (50.0)

2.22 (1.30)

4 (50.0)

2.21 (1.73)

0.995

MEDS (% of cohort)

0 to 4 (12.3)

8 (33.3)

1.68 (1.49)

16 (66.7)

1.69 (1.40)

0.978

5 to 7 (27.2)

18 (34.0)

1.75 (1.67)

35 (66.0)

1.95 (1.74)

0.490

8 to 12 (54.3)

40 (37.7)

1.70 (1.53)

66 (62.3)

1.77 (1.76)

0.680

13 to 18 (6.2)

4 (33.3)

1.60 (1.10)

8 (66.7)

1.41 (1.67)

0.645

PIRO (% of cohort) 0 to 9a (23.1)

12 (26.7)

1.71 (1.41)

33 (73.3)

1.58 (1.53)

0.579

10 to 14 (65.1)

53 (41.7)

1.65 (1.52)

74 (58.3)

1.74 (1.59)

0.504

15 to 19 (11.8)

5 (21.7)

2.40 (1.88)

18 (78.3)

2.46 (2.27)

0.951

N – number of patients; MEDS – Mortality in Emergency Department Sepsis; PIRO – Predisposition, Insult and Response in Organ Failure; SOFA – Sequential (Sepsis-related) Organ Failure Assessment.

a PIRO scores 0 to 4 was combined with PIRO scores 5 to 9 due to too few patients with PIRO scores of 0 to 4.

attributable to the conventional view of widespread tissue hypoxia. In the post-resuscitation stage where systemic perfusion has been largely restored, mechanisms such as microcirculatory and mitochondrial dis- tress syndrome, beta-adrenergic stimulation and reduced lactate clear- ance might come to the fore in driving hyperlactatemia. Most of the evidence supporting beta-adrenergic driven lactate production in septic patients are mainly involving ICU cohorts whereby patients have al- ready received Initial resuscitation prior to ICU admission [8-10]. An in- ference from this postulation is that initial efforts to improve circulatory function in early stages of sepsis should remain a priority, and later management should be geared towards improving microcirculatory function via timely source control to reduce inflammatory host re- sponse against infection and improve microcirculatory oxygen delivery, modulating beta-adrenergic response and monitoring for hepatic dys- function. Thus, the use of agents such as dobutamine, nitric oxide do- nors and acetylcholine [29-31] to improve microvascular perfusion may benefit septic patients and could be the focus of future research.

Although univariate analysis did not show any significant associa- tion between long-term beta-blocker therapy and 28-day all-cause mortality, there was a trend towards lower 28-day all-cause mortality among patients on long term beta-blocker therapy despite them having a higher ICU admission rate, which supports previous evidence demon- strating survival advantage in patients on chronic beta-blocker therapy [13]. Hence, future prospective studies specifically designed to investi- gate the effect of chronic beta-blocker therapy on mortality in sepsis would be needed for further clarification.

Our key findings are in contrast to the results reported by a previous study [12]. We propose several reasons to explain this discrepancy. This study was designed with limitations of the study by Contenti et al. in mind. We included patients into our study if there was clinical suspicion of sepsis at the ED rather than based on final coded diagnosis, which likely introduced selection bias. Although some patients in our cohort received a diagnosis other than sepsis at discharge, we consider this in- terpretation to be more clinically relevant in the context of the ED, which often receives patients in early stages of disease where definite diagnosis of sepsis is not always apparent until later.

Common causes of lactic acidosis that could confound the primary outcome were not deliberated by previous studies – these include met- formin therapy, chronic liver disease, malignancy and diabetes mellitus [11]. We reduced potential confounding by these factors through exclusion of patients on metformin therapy and those with chronic liver disease, and by adjusting for other comorbidities such as renal in- sufficiency, diabetes mellitus and malignancy using regression analyses. We also excluded patients who received beta-agonist therapy prior to sampling of serum lactate, as it has been shown to cause lactic acidosis [32], which would not have been possible in a retrospective design.

Cohorts were also defined differently. Our cohort was selected using the qSOFA score, a new recommendation from the Sepsis-3 workgroup constructed for use outside ICU settings to identify pa- tients who are at higher risk of poorer outcomes including the need for ICU management and death [1], rather than the previous diagnostic criteria for sepsis [33]. The qSOFA score comprises three components, all of which are objective measures that can be easily obtained by the bedside. Although sepsis is currently defined as an increase in the SOFA score by two or more from baseline in a patient with suspected infection, it is frequently difficult to achieve this in a timely manner in the ED due to requirement of laboratory investigations. Therefore, the qSOFA score could be the tool of choice for rapid identification and risk stratification of ED patients with suspected sepsis. Although current evidence suggests that a qSOFA score of 2 or more demonstrates similar Predictive validity as the SOFA score in settings outside ICU [34], we included patients with qSOFA score of 1 and above in order to increase the spectrum of subjects with infection and sepsis not restricted only to those with higher mortality risk, allowing results to be generalized to a wider range of patients that we commonly encounter in the ED.

Limitations

Our study has its limitations. Beta-blockers are used as anti- hypertensive agents. Hence, patients on chronic beta-blocker therapy may be recruited based on the single criterion of having a systolic blood pressure <= 100 mmHg in the qSOFA score. The hypotension could have been a manifestation of ongoing sepsis, the effect of chronic beta-blocker therapy or a combination of both. To adjust for selection bias, we identified patients who were recruited solely based on the sys- tolic hypotension criterion and adjusted the systolic blood pressure based on data from meta-analyses reporting the blood pressure lower- ing effect of beta-blockers [27,28]. Among these patients, those who failed to meet the hypotension criteria after adjustment were excluded from our post-hoc analysis. Primary and secondary outcomes were not significantly altered after exclusion of just one patient recruited based on the hypotension criterion alone.

Though the use of an inclusion criteria of qSOFA score of 1 and above enabled us to capture a wider spectrum of patients with infection, it pre- dictably led to overall lower severity of sepsis in our sample population, reflected by lower SOFA scores, ICU admissions, 28-day all-cause mortality rates, and higher initial mean arterial pressure. Despite this, stratification using SOFA, MEDS and PIRO scores showed that chronic beta-blocker use did not significantly affect serum lactate levels among patients in our cohort with higher risk.

Table 3

Linear regression analysis.

Independent variables

Univariate analysis

Multivariate analysisa

Coefficients (95% CI)

P value

Coefficients (95% CI)

P value

Female gender

0.908 (0.789 to 1.046)

0.180

Age

1.002 (0.996 to 1.007)

0.581

Race

0.452

  • Chinese

Reference group

  • Malay

0.885 (0.748 to 1.047)

0.152

  • Indian

0.986 (0.754 to 1.290)

0.920

  • Others

1.147 (0.733 to 1.797)

0.546

Nursing home residence

1.364 (0.994 to 1.872)

0.055

Mean arterial pressure

0.997 (0.994 to 1.001)

0.137

Heart rate

1.004 (1.000 to 1.008)

0.027

1.004 (1.000 to 1.008)

0.039

Glasgow Coma Score

0.032

  • 13 to 15

Reference group

  • 9 to 12

1.309 (1.023 to 1.675)

0.032

  • 3 to 8

1.500 (0.919 to 2.451)

0.105

Chronic beta-blocker therapy

0.882 (0.756 to 1.029)

0.110

Hypertension

0.942 (0.807 to 1.099)

0.444

Diabetes mellitus

0.933 (0.799 to 1.089)

0.376

Dyslipidemia

0.895 (0.777 to 1.030)

0.121

Ischemic heart disease

1.010 (0.861 to 1.184)

0.906

Renal impairment

0.904 (0.777 to 1.052)

0.190

Malignancy

0.941 (0.797 to 1.110)

0.468

Principal source of infection

0.267

  • Respiratory

Reference group

  • Genitourinary

1.213 (0.985 to 1.494)

0.069

  • Gastrointestinal

1.170 (0.833 to 1.644)

0.363

  • Hepatobiliary

1.138 (0.810 to 1.599)

0.455

1.220 (0.960 to 1.549)

0.103

  • Undifferentiated

1.012 (0.671 to 1.526)

0.954

  • Others

1.272 (0.996 to 1.624)

0.053

qSOFA

0.022

  • 1

Reference group

  • 2

1.168 (0.987 to 1.382)

0.071

  • 3

1.511 (1.063 to 2.148)

0.022

SOFA

0.177

  • 0 to 6

Reference group

  • 7 to 9

1.275 (0.895 to 1.817)

0.177

MEDS

0.442

  • 0 to 4

Reference group

  • 5 to 7

1.112 (0.873 to 1.416)

0.388

  • 8 to 12

1.033 (0.827 to 1.290)

0.774

  • 13 to 18

0.868 (0.613 to 1.229)

0.422

PIRO

  • 0 to 9b

Reference group

0.002

Reference group

  • 10 to 14

1.055 (0.894 to 1.246)

0.525

1.018 (0.866 to 1.197)

0.826

  • 15 to 19

1.518 (1.188 to 1.940)

0.001

1.335 (1.050 to 1.698)

0.019

ICU admission

1.480 (1.208 to 1.815)

b0.001

1.386 (1.126 to 1.707)

0.002

28-day all-cause mortality

1.382 (1.088 to 1.756)

0.008

ICU – intensive care unit; MEDS – Mortality in Emergency Department Sepsis; PIRO – Predisposition, Insult and Response in Organ Failure; qSOFA – quick Sequential (Sepsis-related) Organ Failure Assessment; SOFA – Sequential (Sepsis-related) Organ Failure Assessment.

a Independent variables with p-values b.10 in the univariate linear regression analysis were included for multivariate linear regression analysis.

b PIRO scores 0 to 4 was combined with PIRO scores 5 to 9 due to too few patients with PIRO scores of 0 to 4.

Aside from the initial serum lactate, subsequent serial lactate measurements and calculated lactate clearance may be important pa- rameters in the management of septic patients, which might be affected by chronic beta-blocker use. Our study did not aim to evaluate these pa- rameters, as only the initial serum lactate would be used in a timely fashion during the triage of a septic patient in the ED, with serial mea- surements being done later on Inpatient units.

It might also be difficult to generalize our results in other healthcare settings as this study was conducted in a single center and participants

Table 4

ICU admission and 28-day all-cause mortality rates of comparator groups.

Variables, n (%) Beta-blockers No beta-blockers P value

were recruited only when a study group member is on site, usually in the day on weekdays. Even though the duration of the study was 18 months from first to last patient recruited, we did not have study team member on-site daily due to clinical and academic commitments. The overall effective recruitment period (when a study team member was present and actively screening) was 146 days. Although conve- nience sampling was used, periods of recruitment were not self- determined but were due to external influences, hence the risk of bias is likely to be low. Additionally, the limitation of patient recruit- ment to office hours and not specific periods would also not signifi- cantly affect incidence or severity of sepsis. The patients included in our study were limited to patients triaged to the immediate and urgent care areas. On average, our emergency department sees 60 to 80 PAC 1 and PAC 2 patients per shift. During the presence of

ICU admission

13 (18.6)

12 (9.6)

0.072

study team members, all patients triaged to PAC 1 and PAC 2 would

28-day all-cause mortality

3 (4.3)

15 (12.0)

0.074

be screened for eligibility based on their triage notes in our elec-

ICU – intensive care unit; n –

number of patients.

tronic medical records system.

  1. Conclusion

In conclusion, long-term beta-blocker therapy did not significantly affect initial serum venous lactate concentration when patients first present to the ED with suspected sepsis. Physicians using serum lactate for diagnosis, risk stratification and further management should not take chronic beta-blocker use into consideration when interpreting serum lactate.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2019.12.046.

Author contributions

JZWC, KSL, MTC, WSK conceived the study, designed the trial, and obtained research funding.

JZWC, MTC, WSK supervised the conduct of the trial and data collection.

JZWC, JHT, AJYN, ZO, XZ undertook recruitment of patients and man- aged the data, including quality control.

JZWC, JHT, MTC, WSK provided statistical advice on study design and analyzed the data.

All authors contributed substantially to the writing of the manu- script, read and approved the final manuscript.

Award at scientific meeting

Australasian College for Emergency Medicine, 2018 Annual Scientific Meeting, Perth, Western Australia. Best paper by a trainee for Beta- Blockers’ Effect on Levels of Lactate (BeLLa) awarded to Dr. Jonathan Zhao Wang Chan on 22 November 2018.

Declaration of competing interest

  1. No conflict of interests JHT, KSL, AJYN, ZO, XZ, MTC and WSK report no conflict of interests.
  2. Grant money for investigator-initiated research

JZWC reports grant money to NUHS to conduct this research con- ceived and written by JZWC from NUHS.

Acknowledgements

The authors would like to thank our colleagues in the Emergency Medicine Department of National University Hospital, Singapore, for their assistance and support in the study.

Source of funding

This study is funded by the National University Health Systems (NUHS) Junior Pitch for Funds Grant (grant reference number JPFF-16- 2-CZWJ).

References

  1. Singer M, Deutschman CS, Seymour CW, et al. The third international consensus def- initions for sepsis and septic shock (Sepsis-3). JAMA 2016;315(8):801.
  2. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Intensive Care Med 2017;43(3):304-77.
  3. Casserly B, Phillips GS, Schorr C, et al. Lactate measurements in sepsis-induced tissue hypoperfusion. Crit Care Med 2015;43(3):567-73.
  4. Garcia-Alvarez M, Marik P, Bellomo R. Sepsis-associated hyperlactatemia. Crit Care 2014;18(5):503.
  5. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014;371:2309-19.
  6. Ince C. The microcirculation is the motor of sepsis. Crit Care 2005;9(4):S13.
  7. Levraut J, Ciebiera JP, Chave S, et al. Mild hyperlactatemia in stable septic patients is due to impaired lactate clearance rather than overproduction. Am J Respir Crit Care Med 1998;157(4):1021-6.
  8. Wutrich Y, Barraud D, Conrad M, et al. Early increase in arterial lactate concentration under epinephrine infusion is associated with a better prognosis during shock. Shock 2010;34(1):4-9.
  9. Morelli A, Donati A, Ertmer C, et al. Microvascular effects of heart Rate control with esmolol in patients with septic shock: a pilot study. Crit Care Med 2013;41(9): 2162-8.
  10. Morelli A, Ertmer C, Westphal M, et al. Effect of Heart rate control with esmolol on hemodynamic and clinical outcomes in patients with septic shock: a randomized clinical trial. JAMA 2013;310(16):1683-91.
  11. Suetrong B, Walley KR. Lactic acidosis in sepsis: it’s not all anaerobic. Chest 2016;149

(1):252-61.

  1. Contenti J, Occelli C, Corraze H, et al. Long-term ?-blocker therapy decreases blood lactate concentration in severely septic patients. Crit Care Med 2015;43(12): 2616-22.
  2. Macchia A, Romero M, Comignani PD, et al. Previous prescription of ?-blockers is as- sociated with reduced mortality among patients hospitalized in intensive care units for sepsis. Crit Care Med 2012;40(10):2768-72.
  3. Stawicki SP, Gerlach AT. Polypharmacy and Medication errors: stop, listen, look, and analyze. Opus 12 Scientist 2009;3:6-10.
  4. Evans DC, Cook CH, Christy JM, et al. Comorbidity-polypharmacy scoring facilitates outcome prediction in older trauma patients. J Am Geriatr Soc 2012;60(8):1465-70.
  5. Kuan WS, Ibrahim I, Leong BS, et al. Emergency department management of sepsis patients: a randomized, goal-oriented, noninvasive sepsis trial. Ann Emerg Med 2016;67(3):367-78.
  6. Fuchs C, Wauschkuhn S, Scheer C, et al. Continuing chronic beta-blockade in the acute phase of severe sepsis and septic shock is associated with decreased mortality rates up to 90 days. Brit J Anaesth 2017;119(4):616-25.
  7. Pandharipande PP, Shintani AK, Hagerman HE, et al. Derivation and validation of SpO2/FiO2 ratio to impute for PaO2/FiO2 ratio in the respiratory component of the Sequential Organ Failure Assessment score. Crit Care Med 2009;37(4):1317.
  8. Shapiro NI, Wolfe RE, Moore RB, et al. Mortality in Emergency Department Sepsis (MEDS) score: a prospectively derived and validated clinical prediction rule. Crit Care Med 2003;31(3):670-5.
  9. Vorwerk C, Loryman B, Coats TJ, et al. Prediction of mortality in adult emergency de- partment patients with sepsis. Emerg Med J 2009;26(4):254-8.
  10. Howell MD, Talmor D, Schuetz P, et al. Proof of principle: the predisposition, infec- tion, response, organ failure sepsis staging system. Crit Care Med 2011;39(2):322-7.
  11. Vincent JL, De Mendonca A, Cantraine F, et al. Use of the SOFA score to assess the in- cidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med 1998;26(11):1793-800.
  12. Lexicomp. Atenolol: drug information. UpToDate. Available from: https://www. uptodate.com/contents/atenolol-drug-information; 2019, Accessed date: 26 May

2019.

  1. Lexicomp. Bisoprolol: Drug information. UpToDate; 2019 Available from: https:// www.uptodate.com/contents/bisoprolol-drug-information, Accessed date: 26 May 2019.
  2. Lexicomp. Carvedilol: drug information. UpToDate. Available from: https://www. uptodate.com/contents/carvedilol-drug-information; 2019, Accessed date: 26 May

2019.

  1. Lexicomp. Metoprolol: drug information. UpToDate. Available from: https://www. uptodate.com/contents/metoprolol-drug-information; 2019, Accessed date: 26

May 2019.

  1. Wong GWK, Laugerotte A, Wright JM. Blood pressure lowering efficacy of dual alpha and Beta blockers for primary hypertension. Cochrane Database Syst Rev 2015(8).
  2. Wong GWK, Boyda HN, Wright JM. Blood pressure lowering efficacy of beta-1 selec- tive beta blockers for primary hypertension. Cochrane Database Syst Rev 2016;3(3): CD007451.
  3. De Backer D, Creteur J, Dubois MJ, et al. The effects of dobutamine on microcircula- tory alterations in patients with septic shock are independent of its systemic effects. Crit Care Med 2006;34(2):403-8.
  4. Spronk PE, Ince C, Gardien MJ, et al. Nitroglycerin in septic shock after intravascular volume resuscitation. Lancet 2002;360(9343):1395-6.
  5. De Backer D, Creteur J, Preiser JC, et al. Microvascular blood flow is altered in patients with sepsis. Am J Respir Crit Care Med 2002;166(1):98-104.
  6. Kraut JA, Madias NE. Lactic acidosis. N Engl J Med 2014;371(24):2309-19.
  7. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ failure and guide- lines for the use of innovative therapies in sepsis. Chest 1992;101(6):1644-55.
  8. Seymour CW, Liu VX, Iwashyna TJ, et al. Assessment of clinical criteria for sepsis: for the third international consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016;315(8):762-74.

Leave a Reply

Your email address will not be published. Required fields are marked *