Lactate Clearance and Normalization and Prolonged

Organ Dysfunction in Pediatric Sepsis

Halden F. Scott, MD1,2, Lina Brou, MPH2, Sara J. Deakyne, MPH1, Diane L. Fairclough, DrPH3,4,
Allison Kempe, MD, MPH1,4,5, and Lalit Bajaj, MD, MPH1,2

Objectives To evaluate whether lactate clearance and normalization during emergency care of pediatric sepsis is
associated with lower rates of persistent organ dysfunction.
Study design This was a prospective cohort study of 77 children <18 years of age in the emergency department
with infection and acute organ dysfunction per consensus definitions. In consented patients, lactate was measured
2 and/or 4 hours after an initial lactate; persistent organ dysfunction was assessed through laboratory and physician
evaluation at 48 hours. A decrease of $10% from initial to final level was considered lactate clearance; a final level
<2 mmol/L was considered lactate normalization. Relative risk (RR) with 95% CIs, adjusted in a log-binomial model,
was used to evaluate associations between lactate clearance/normalization and organ dysfunction.
Results Lactate normalized in 62 (81%) patients and cleared in 70 (91%). The primary outcome, persistent 48-hour
organ dysfunction, was present in 32 (42%). Lactate normalization was associated with decreased risk of persistent
organ dysfunction (RR 0.46, 0.29-0.73; adjusted RR 0.47, 0.29-0.78); lactate clearance was not (RR 0.70, 0.35-1.41;
adjusted RR 0.75, 0.38-1.50). The association between lactate normalization and decreased risk of persistent organ
dysfunction was retained in the subgroups with initial lactate $2 mmol/L and hypotension.
Conclusions In children with sepsis and organ dysfunction, lactate normalization within 4 hours was associated
with decreased persistent organ dysfunction. Serial lactate level measurement may provide a useful prognostic tool
during the first hours of resuscitation in pediatric sepsis. (J Pediatr 2015;-:---).

M
ore than 75 000 US children develop severe sepsis yearly, with US pediatric mortality rates of 10%-20%.1,2 Failure to
recognize and adequately resuscitate shock have been identified as causes of preventable morbidity.3,4 The ability to
initially identify the highest-risk children and to objectively monitor the progression of disease and success of therapy
at bedside is critical to the quality of sepsis care.
Elevated lactate, a byproduct of anaerobic metabolism, is associated with poor prognosis in sepsis, and changes in real time in
response to pathophysiology and resuscitation.5-8 Measuring and re-measuring lactate levels comprise 2 of 7 surviving sepsis
bundle elements whose implementation in adults has decreased mortality.9,10 A single elevated serum lactate level at the start of
emergency department (ED) care in pediatric sepsis is associated with a 5-fold increase in organ dysfunction risk, but no studies
have evaluated lactate clearance or normalization in early pediatric sepsis.6 Thus, this study sought to test whether a decrease in
serum lactate level during the first 4 hours of care is associated with lower rates of persistent organ dysfunction at 48 hours,
among pediatric ED patients with infection and acute organ dysfunction. The hypothesis was that patients with lactate
clearance and normalization would have lower rates of persistent organ dysfunction at 48 hours.

Methods
This was a prospective cohort study conducted at Children’s Hospital Colorado,
an academic pediatric referral center that treats >70 000 patients yearly in the ED.
The ED uses a sepsis resuscitation system, in which ED clinicians may activate a
coordinated, standardized, resuscitation response via page when they clinically From the 1Children’s Hospital Colorado; 2Section of
Emergency Medicine, Department of Pediatrics,
determine that a patient has suspected sepsis with decreased mental status or University of Colorado School of Medicine; 3Colorado
perfusion. Measuring a lactate level is recommended but not required in cases School of Public Health; 4Adult and Child Center for
Outcomes Research and Delivery Science, University of
of suspected septic shock. Internal continuous quality improvement data show Colorado and Children’s Hospital Colorado; and
5
Department of Pediatrics, University of Colorado School
that lactate is tested in 60% of cases of suspected sepsis and 75% of cases of septic of Medicine, Aurora, CO
Supported by the Thrasher Research Fund, Early Career
Award (11363 [PI: H.S.]). The REDCap database used in
this study is supported by Colorado Clinical & Transla-
tional Sciences Institute with the Development and
ED Emergency department Informatics Service Center by the National Institutes of
ICU Intensive care unit Health/National Center for Research Resources (UL1
TR001082). The authors declare no conflicts of interest.
PELOD Pediatric Logistic Organ Dysfunction
REDCap Research Electronic Data Capture 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All
RR Relative risk rights reserved.
http://dx.doi.org/10.1016/j.jpeds.2015.11.071

1
THE JOURNAL OF PEDIATRICS  www.jpeds.com
shock with hypotension.11 Children who presented for care
to the ED were eligible if they were <18 years old, had a lactate
level measured by their treating physician, had suspected
infection as identified by their treating physician, and met
organ dysfunction criteria in the ED. Published consensus
definitions for organ dysfunction were used (Table I;
available at www.jpeds.com).12 Children with inborn errors
of metabolism and children whose ED care occurred
between 1:00 a.m. and 7:00 a.m., which was outside of
enrollment hours, were excluded. Activation of the sepsis
resuscitation system was not required for enrollment in the
study. This study was approved by the Colorado Multiple
Institutional Review Board.
Research assistants identified potential subjects who had a
lactate level measured during their ED care. Treating
physicians were then asked to complete a screening form to
identify whether infection was suspected and to assess
inclusion and exclusion criteria. Patients meeting inclusion
criteria were then approached for written informed consent
from the parents or guardians of the subjects, and when
appropriate, assent was obtained from the children
themselves.
After consenting patients were enrolled, serial venous
lactate levels were measured. Attempts to collect lactate levels
occurred 2 and 4 hours after the original (nonstudy) clinical
lactate level was collected. Although both the 2- and 4-hour
measurements were preferred, a collection at a minimum
of one of these times was considered acceptable for
continuation in the study. This approach reflects Surviving
Sepsis Guidelines that recommend a repeat lactate
measurement during the first 6 hours of care but do not
specify a precise timepoint for collection, reflecting clinical
realities of acute resuscitation and challenges of vascular ac-
cess.9,13 Consistent with prior studies, venous lactate levels
were collected by nurses at the same time as collection of
other clinical samples, using standard institutional
procedures for collection of blood samples, and tourniquets
were permitted.6,13,14 Blood gas syringes for measurement of
lactate were immediately placed on ice and levels measured in
the Children’s Hospital Colorado laboratory, an accredited
laboratory with quality control procedures, using the
Siemens Rapidlab 1265 (Siemens Healthcare Diagnostics
Inc, Tarrytown, New York).
The consensus definitions for organ dysfunction use
clinical, laboratory, and physical examination findings.12
Laboratory tests required to assess organ dysfunction
were collected 48 hours after the initial lactate. Physical
examination findings required to assess organ dysfunction
were recorded on a data collection form by the patient’s
attending physician at 48 hours. Patients who were
discharged from medical care prior to 48 hours had clinical
and laboratory assessments for organ dysfunction completed
immediately prior to discharge. Additional clinical data
including vital signs, laboratory results, hospital course,
and treatments were obtained via structured chart
abstraction. Study data were collected and managed using
Research Electronic Data Capture (REDCap), hosted at the
2 Scott et al
Volume -
University of Colorado. REDCap is a secure, web-based
application designed to support data capture for research
studies.15 Data were then exported for analysis from REDCap
to SAS (SAS Institute, Cary, North Carolina).
Clinicians were blinded to laboratory results collected for
the study and treatment was performed according to
standard of care.
Measures
When both 2- and 4-hour lactate levels were both
collected, the 4-hour value was used to determine lactate
clearance and lactate normalization status. Lactate clearance
was defined according to published definitions as a
decrease in serum lactate of $10% between first and
second measurement in patients with initial lactate
$2 mmol/L (18 mg/dL), or a second measurement
<2 mmol/L (18 mg/dL) in patients with initial lactate
<2 mmol/L (18 mg/dL).13,16 Lactate normalization was a
second lactate level that was <2 mmol/L (18 mg/dL).16 The
primary outcome measure was at least one organ dysfunction
persisting $48 hours after triage time.
Organ dysfunction was defined using international
pediatric sepsis consensus definitions (Table I).12 In this
study, elevated lactate was not considered part of the
outcome of organ dysfunction. The Pediatric Logistic
Organ Dysfunction (PELOD) score, a validated measure of
organ dysfunction and mortality risk in children, was
calculated for the first hospital day to assess severity of
presenting illness. PELOD is a score that uses weighted
measures of organ dysfunction to predict mortality among
pediatric intensive care unit (ICU) patients.17,18
Statistical Analyses
Data were summarized using standard descriptive statistics.
Continuous outcomes were compared with Wilcoxon
rank-sum, including ventilator-days, vasopressor-days,
organ-dysfunction-days, and length of stay in hospital and
intensive care. Dichotomous outcomes such as mortality,
vasopressor requirement, intubation, and disposition to the
ICU were compared with the Fisher exact statistic.
The sample size was determined based on the planned anal-
ysis for the comparison of presence of 48-hour organ dysfunc-
tion between lactate clearance and nonclearance groups,
using the Fisher exact statistic. Based on pediatric and adult
studies of lactate in sepsis, it was estimated that 33% of sub-
jects would not have lactate clearance and that 40% of those
without lactate clearance would have 48-hour organ dysfunc-
tion; 5% of the group with lactate clearance would have
48-hour organ dysfunction.6,13,19 With a sample size of 78
total patients, this would provide 80% power to detect this
difference with a 2-tailed alpha of 0.05.
To assess potential confounding effects of select variables
on the association of lactate normalization with prolonged
organ dysfunction, an adjusted relative risk (RR) was
calculated using the log-binomial model with backward
elimination. Covariates were selected based on biological
plausibility, prior studies, and univariate association with
- 2015 ORIGINAL ARTICLES

Table III. Subject characteristics, by lactate normalization or non-normalization status

Characteristics Lactate normalization (n = 62) Lactate non-normalization (n = 15) P value
Baseline characteristics
Age (y) 12.0 [6.0, 14.0] 8.0 [2.0, 16.0] .64
Chronic medical condition 40 (65%) 12 (80%) .76
Neuromuscular 22 (35%) 6 (40%)
Pulmonary 14 (23%) 4 (27%)
Endocrine/metabolic 8 (13%) 1 (7%)
Oncologic 7 (11%) 2 (13%)
Gastrointestinal/liver 6 (10%) 4 (27%)
Renal 4 (6%) 1 (7%)
Hematologic 2 (3%) 1 (7%)
Cardiac 2 (3%) 0 (0%)
Central line present 11 (18%) 4 (27%) .43
Initial lactate level (mmol/L) 2.0 [1.3, 2.6] 3.6 [2.9, 4.4] .0001*
Initial PELOD score 10.0 [10.0, 11.0] 12.0 [2.0, 20.0] .11
Number of ED organ system dysfunctions 1.0 [1.0, 2.0] 2.0 [1.0, 3.0] .0003*
Hypotension prior to enrollment 34 (55%) 8 (53%) .99
Treatment characteristics
ED sepsis resuscitation activation 26 (42%) 6 (40%) .99
Time to antibiotics (minutes) 92.5 [48.5, 178.0] 122.0 [81.0, 188.0] .37
Total volume of isotonic fluid/patient weight delivered by 4 h (mL/kg) 40.0 [21.5, 52.5] 43.8 [28.7, 63.9] .36
Vasoactive agents between h 0 and 4 16 (26%) 5 (33%) .54
Clinical outcomes
Hospital length of stay (d) 4.9 [2.9, 8.0] 9.0 [5.7, 16.0] .02*
PICU length of stay (d) 1.6 [0.0, 3.0] 4.0 [1.4, 10.9] .01*
Positive-pressure ventilation d 0.0 [0.0, 1.8] 0.0 [0.0, 4.2] .38
Vasoactive agent d 0.0 [0.0, 1.0] 0.0 [0.0, 3.0] .09
Number of organ systems dysfunctional at 48 h 0.0 [0.0, 1.0] 1.0 [0.0, 3.0] .002*
In-hospital mortality 1 (2%) 1 (7%) .35
Primary source of infection identified 42 (68%) 12 (80%) .53
Pneumonia 21 (34%) 3 (20%)
Urosepsis 7 (11%) 2 (13%)
Abdominal 7 (11%) 2 (13%)
Blood 6 (10%) 9 (60%)
Toxic shock 3 (5%) 0 (0%)
Skin/soft tissue/musculoskeletal 2 (3%) 1 (7%)
Meningitis 0 (0%) 1 (7%)
Microbiologic infection identified 33 (53%) 11 (73%) .25
Viral PCR positive 18 (29%) 2 (13%)
Bacteria culture positive 16 (26%) 8 (53%)
Fungal culture positive 2 (3%) 0 (0%)
Malaria positive 0 (0%) 1 (7%)

PICU, pediatric ICU; PCR, polymerase chain reaction.

Continuous and ordinal variables reported as median [IQR]; categorical variables reported with (percentage). P values calculated with Fisher exact statistic for categorical variables and Wilcoxon
rank-sum for continuous variables.
All microbiologic sources and chronic conditions in all subjects are listed, with some subjects having multiple sources and conditions.
*Indicates significance to the level of P < .05.

the outcome. Covariates were retained in the model that
changed the base model RR by 10% or greater.
Several subgroup analyses were planned a priori, including
of the subgroup of patients with an initial lactate level
$2.0 mmol/L (18 mg/dL). In order to test whether lactate
clearance and normalization added prognostic value to clinical
risk-stratification, subgroup analyses of subjects with and
without ED hypotension, and subjects with and without ED
sepsis resuscitation activations were planned a priori. Activation
of a sepsis resuscitation served as a marker for clinical risk
assessment, representing the highest level of resources and
care for septic shock available to clinicians at their discretion.
Results
Of 600 patients with an initial clinical lactate level collected
by clinicians, 110 were eligible for inclusion, and 85 subjects
were enrolled between April 2012 and September 2014
Lactate Clearance and Normalization and Prolonged Organ Dysfunction in Pediatric Sepsis
(Figure 1; available at www.jpeds.com). In 8 enrolled
patients, a second lactate level was unable to be collected
before hour 4 because of intravenous access and active
resuscitation procedures; these patients were excluded from
analysis (Table II; available at www.jpeds.com). In 1 of
these 8 patients, this was noted after study enrollment had
closed, resulting in a population of 77 analyzable subjects,
instead of the intended 78.
Lactate normalized in 62 (81%) patients and cleared in
70 (91%). Study population characteristics, including
differences between subjects with and without lactate
normalization are shown in Table III, and differences
between subjects with and without lactate clearance are
shown in Table IV (available at www.jpeds.com). The
primary outcome, organ dysfunction at 48 hours, was
present in 32 (42%) patients. Consistent with the study
hypothesis, lactate normalization was significantly
associated with decreased risk of persistent organ
3
THE JOURNAL OF PEDIATRICS  www.jpeds.com
Figure 2. Proportion of subjects with persistent organ dysfunction at 48 hours. The number of subjects with persistent organ
dysfunction out of the total number of subjects in a given group is indicated with a fraction. Fisher exact test used to calculate
P values. *Indicates significant associations.
dysfunction, with an unadjusted RR of 0.46 (95% CI:
0.29-0.73) (Figure 2). Although lower rates of organ
dysfunction were seen in patients with lactate clearance,
these differences did not reach statistical significance, with
unadjusted RR of 0.70 (0.35-1.41), which did not support
the study hypothesis. These associations were similar in the
subgroup of patients with an elevated initial lactate
(Figure 2).
The initial variables included in the log-binomial model
were: age, PELOD score, chronic medical condition, time
to antibiotics, and volume of isotonic fluid delivered/patient
weight. Inclusion of initial lactate in the model was
considered, but it demonstrated collinearity with the
predictors, and models including initial lactate did not satisfy
maximum likelihood convergence criteria so it was not
included. After iterative removal of variables that did not
change the base RR by at least 10%, the final model included:
age, initial PELOD score, chronic medical condition, as well
as lactate clearance or lactate normalization. The adjusted RR
for the association between lactate normalization and pro-
longed organ dysfunction was 0.47 (95% CI 0.29-0.78). The
adjusted RR for lactate clearance was 0.75 (95% CI 0.38-
1.50).
Among patients with ED hypotension, failure to normalize
lactate was associated with a significantly increased risk of
persistent organ dysfunction (Figure 3). Among patients
without sepsis activations, failure to normalize lactate was
also associated with an increased risk of persistent organ
dysfunction (Figure 3).
4 Scott et al
Volume -
Of the 48 patients who had both the 2- and 4-hour lactate
level tested, 43 (89.6%) were concordant for lactate
normalization status between the 2 times, with a kappa of
0.73 (0.51-0.95). There were only 5 (6.5%) patients in
whom lactate normalization status differed between 2
and 4 hours; with so few patients driving the difference, the
2- and 4-hour lactate normalization groups were not
analyzed separately. In 4 of these 5 cases where the 2- and
4-hour lactate normalization status differed, the 4-hour
lactate level correctly predicted 48-hour organ dysfunction.
The median ED length of stay in the study population was
hours (IQR 2.4-5.8 hours).
Median initial lactate levels were similar when collected
from central and peripheral venous catheters, overall and
when grouped according to initial PELOD score (Table V;
available at www.jpeds.com). In addition to starting at a
lower median lactate level, there was a larger median
absolute and relative decrease in lactate level between the
first and final measurements in the lactate normalization
group than in the lactate non-normalization group
(Table VI; available at www.jpeds.com).
Discussion
This study provides evidence that serial lactate measurements
may add useful prognostic information in the first hours of
pediatric sepsis care in a non-ICU setting. These are critical
hours when lapses in care, and subsequent preventable
morbidity, are frequently due to failure in recognizing
- 2015 ORIGINAL ARTICLES

Figure 3. Subgroup analyses: lactate normalization and percent of subjects with persistent organ dysfunction at 48 hours in
subjects with and without ED hypotension and with and without an ED sepsis resuscitation activation. A sepsis activation
indicates that a clinician has identified the patient as requiring emergent critical resuscitation. The number of subjects with
persistent organ dysfunction out of the total number of subjects in a given group is indicated with a fraction. Fisher exact test
used to calculate P values. *Indicates significant associations.

severity of illness.3,4 It is also a time when advanced, invasive
monitoring techniques may be physically unavailable or not
yet in place. The current study demonstrated that lactate
normalization is associated with a decreased risk of
prolonged organ dysfunction at 48 hours in ED patients
with suspected sepsis and organ dysfunction.
In addition, this study suggests that lactate normalization
remains associated with prolonged organ dysfunction among
patients in clinically relevant subgroups. In the cohort of
patients considered lower-acuity by physicians (in whom
physicians did not activate the sepsis resuscitation system),
failure to normalize lactate was significantly associated with
persistent organ dysfunction, demonstrating the potential
for lactate clearance measurements to identify patients not
identified by physicians at this site for sepsis resuscitation.
Similarly, failure to normalize lactate conferred significant
additional risk in patients with hypotension in the ED.
Patients with hypotension and failure to normalize lactate
were the highest-risk cohort in this study, consistent with
adult sepsis studies.20 As a high-morbidity cohort, this is an
important group to identify clinically and could be
well-suited for future prospective research trials.
These findings are consistent with previous studies of
lactate. Lactate clearance has been associated with improved
outcomes in wide-ranging populations and conditions
including adult sepsis, trauma, and pediatric cardiac
surgery.13,21-23 The current study lends insight into the first
hours of emergency pediatric sepsis care. Normalization of
lactate by 4 hours of arrival may be more easily adopted by
clinicians than levels over the first 6-24 hours after admission
or area under the lactate curve, which previous studies have
used.7,8,23
Recent studies converge to highlight the critical need for
objective measures with which to diagnose pediatric septic
shock and monitor progression of disease and response to
therapy. Research sheds doubt on the ability of the physical
examination and Systemic Inflammatory Response
Syndrome vital signs to diagnose septic shock and identify
the precise hemodynamic state of a patient.24-26 Concerns
that intravenous fluid resuscitation may cause harm in
certain populations highlight the need for precise tailoring
of fluid resuscitation to each patient.27 Although lactate
normalization does not precisely describe a hemodynamic
state, it is an objective, inexpensive, and minimally invasive
measure capable of informing bedside clinicians in
real-time whether a patient is grossly improving or worsening
in their care. The addition of lactate clearance as a clinical
target of care to a bundle of resuscitation elements
implemented in the care of adults with septic shock reduced
the risk of death; further studies are needed to determine
whether its use would improve outcomes in pediatric septic
shock as well.10
Central venous oxygen saturation monitoring and
ultrasound-based monitoring during pediatric sepsis

Lactate Clearance and Normalization and Prolonged Organ Dysfunction in Pediatric Sepsis 5
THE JOURNAL OF PEDIATRICS  www.jpeds.com
improved outcomes.28,29 Although promising, these tech-
niques require specialized equipment and technical
proficiency. Lactate testing requires little skill and is widely
available, making it more likely to be implemented as an
approach to enhanced monitoring.
In this study, there were not significant differences in
volume of fluid resuscitation between cohorts with and
without lactate normalization, suggesting that normalization
was not only a direct reflection of volume of fluid
delivered. Shock resuscitation depends on tailoring fluid
delivery and vasoactive choice and timing to individual
patient condition, and it may be that the patients who
normalized lactate were receiving therapies better targeted
to their specific hemodynamic state, not simply receiving
more.
Patients included in this study met strict organ
dysfunction definitions in the ED, making them more
severely ill than many patients who benefit from sepsis
treatment but who are not yet demonstrating end-organ
effects. In addition, patients were enrolled after a clinical
lactate specimen was collected. Thus, the study results reflect
the utility of lactate normalization only in a population in
whom physicians are concerned enough to measure a lactate
level. Practices in measuring lactate vary, thus, potentially
adversely affecting external generalizability of the study,
although diagnosis and treatment of sepsis currently depend
upon physician suspicion. The study population is similar to
the populations of previous studies of tertiary pediatric ED
patients with sepsis, including age, mortality rates, infectious
source, and proportions with medical complexity.30-32 The
use of clinical impression of infection as part of the
framework to identify patients with sepsis reflects all
published sepsis treatment guidelines because patients must
receive lifesaving treatment before a microbiological
diagnosis is available.9,33
During study enrollment, an updated PELOD-2 score was
published. Because a major innovation in the new score was
the addition of a lactate level, the original PELOD score was
retained for this study so as to avoid collinearity in the
model.34 The inclusion of lactate in the PELOD-2 score is
in keeping with our findings that lactate levels add additional
prognostic value to existing risk factors.
The power of the study was limited by the size and the fact
that rates of lactate clearance were higher than estimated in a
priori sample size calculations. Sample size calculations were
based on a previous study of lactate in pediatric sepsis,
conducted prior to pediatric sepsis quality improvement
programs, which have improved care, likely resulting in
higher rates of lactate clearance than had been present
historically.6 Compared with published studies of lactate
clearance in adult patients with severe sepsis and septic shock,
this study population had higher rates of lactate clearance,
likely because of lower overall illness severity.5,10 However,
lactate levels reflect multifactorial perfusion and metabolic
processes, and the pathophysiology of sepsis changes with
age, so there may be physiologic reasons that lactate clearance
might be more easily achieved in a pediatric population as
6 Scott et al
Volume -
well. Although in the study there were lower rates of organ
dysfunction in patients with lactate clearance, the association
was not statistically significant; given the lower rates of lactate
clearance seen in the study than used in the a priori power
calculations, a larger sample size might have better powered
the study to determine the significance of this association.
The difference in time to antibiotic, which was longer but
not significantly so in the non-normalization cohort, might
be statistically significant with a larger sample size. Adult
definitions for lactate clearance and normalization were
adopted for this study; a larger study could potentially
discern thresholds with improved test characteristics for
pediatrics.
This is an observational study in which results have
potential to be confounded by the care delivered. However,
measures of the quality of care such as volume of fluid
resuscitation and timeliness of antibiotic delivery were
included in the original binomial model before
refinement, and these variables did not significantly
influence the association between lactate normalization and
outcome.
Sample collection at both 2 and 4 hours was not possible in
all patients. Although research assistants were prompting
sample collection, difficult intravenous access and competing
needs of critical resuscitation prevented collection in many
cases. The characteristics of patients excluded from analysis
because of difficulty obtaining the sample are similar to the
overall study population (Table II). Using samples
collected at 2 or 4 hours in this analysis is consistent with
adult literature, in which a repeat lactate level was
measured at any point within the first 6 hours of care.13,19
The significance of the lactate measurements in this study,
despite the imprecision of sample collection in the first
hours of critical resuscitation, suggests that these
measurements may be effective in a chaotic ED setting.
There was good agreement between a 2- and 4-hour lactate
level. Research samples were collected along with clinical
samples, reflecting that this test is effective when collected
under real-world clinical conditions and is consistent with
prior studies validating that lactate levels are meaningful
and accurate despite tourniquet use, prolonged in vitro
time, and venous sampling.35
Preventable pediatric deaths from sepsis are frequently due
to failure to recognize or aggressively treat a shock state. A
physiologic measure of the progression of disease and success
of therapy has potential to improve sepsis treatment at the
bedside. This study suggests that measurement of serial
lactate levels in the first 4 hours of treatment of pediatric
sepsis may be a useful indicator of prognosis. As research
to better characterize and precisely target therapies in
pediatric sepsis to individual hemodynamic states continues,
future trials might be designed to test the use of serial lactate
measurements to guide therapy. n
We acknowledge the contributions of Kendra Kocher, Kathleen Grice,
and Mimi Goodwin (University of Colorado; funded by the
Thrasher Research Fund Early Career Award [11363]) in training
- 2015
and supervising research assistants, supervising data entry and
compliance, and assisting with manuscript preparation.
Submitted for publication Aug 21, 2015; last revision received Nov 9, 2015;
accepted Nov 24, 2015.
Reprint requests: Halden F. Scott, MD, Children’s Hospital Colorado, 13123 E
16th Ave, B251, Aurora, CO 80045. E-mail: halden.scott@childrenscolorado.
org
References
1. Hartman ME, Linde-Zwirble WT, Angus DC, Watson RS. Trends in the
epidemiology of pediatric severe sepsis. Pediatr Crit Care Med 2013;14:
686-93.
2. Balamuth F, Weiss SL, Neuman MI, Scott H, Brady PW, Paul R, et al.
Pediatric severe sepsis in US children’s hospitals. Pediatr Crit Care
Med 2014;15:798-805.
3. Han YY, Carcillo JA, Dragotta MA, Bills DM, Watson RS,
Westerman ME, et al. Early reversal of pediatric-neonatal septic shock
by community physicians is associated with improved outcome.
Pediatrics 2003;112:793-9.
4. Inwald DP, Tasker RC, Peters MJ, Nadel S. Paediatric Intensive Care
Society Study G. Emergency management of children with severe sepsis
in the United Kingdom: the results of the Paediatric Intensive Care
Society sepsis audit. Arch Dis Child 2009;94:348-53.
5. Nguyen HB, Rivers EP, Knoblich BP, Jacobsen G, Muzzin A, Ressler JA,
et al. Early lactate clearance is associated with improved outcome in
severe sepsis and septic shock. Crit Care Med 2004;32:1637-42.
6. Scott HF, Donoghue AJ, Gaieski DF, Marchese RF, Mistry RD. The
utility of early lactate testing in undifferentiated pediatric systemic
inflammatory response syndrome. Acad Emerg Med 2012;19:1276-80.
7. Kim Y, Ha E-J, Jhang W, Park S. Early blood lactate area as a prognostic
marker in pediatric septic shock. Intensive Care Med 2013;39:1818-23.
8. Jat KR, Jhamb U, Gupta VK. Serum lactate levels as the predictor of
outcome in pediatric septic shock. Indian J Crit Care Med 2011;15:
102-7.
9. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al.
Surviving sepsis campaign: international guidelines for management of
severe sepsis and septic shock: 2012. Crit Care Med 2013;41:580-637.
10. Nguyen HB, Kuan WS, Batech M, Shrikhande P, Mahadevan M, Li CH,
et al. Outcome effectiveness of the severe sepsis resuscitation bundle with
addition of lactate clearance as a bundle item: a multi-national
evaluation. Crit Care 2011;15:R229.
11. Scott HF, Brou L, Deakyne SJ, Bajaj L. Delayed hypotension in pediatric
sepsis: frequency, outcomes and quality of care. American Academy of
Pediatrics National Conference and Exhibition. San Diego, CA; 2014.
12. Goldstein B, Giroir B, Randolph A , International Consensus Conference
on Pediatric Sepsis. International pediatric sepsis consensus conference:
definitions for sepsis and organ dysfunction in pediatrics. Pediatr Crit
Care Med 2005;6:2-8.
13. Arnold RC, Shapiro NI, Jones AE, Schorr C, Pope J, Casner E, et al.
Multicenter study of early lactate clearance as a determinant of survival
in patients with presumed sepsis. Shock 2009;32:35-9.
14. Reed L, Carroll J, Cummings A, Markwell S, Wall J, Duong M. Serum
lactate as a screening tool and predictor of outcome in pediatric patients
presenting to the emergency department with suspected infection.
Pediatr Emerg Care 2013;29:787-91.
15. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research
electronic data capture (REDCap)—a metadata-driven methodology
and workflow process for providing translational research informatics
support. J Biomed Inform 2009;42:377-81.
16. Puskarich MA, Trzeciak S, Shapiro NI, Albers AB, Heffner AC, Kline JA,
et al. Whole blood lactate kinetics in patients undergoing quantitative
resuscitation for severe sepsis and septic shock. Chest 2013;143:1548-53.
Lactate Clearance and Normalization and Prolonged Organ Dysfunction in Pediatric Sepsis
ORIGINAL ARTICLES
17. Leteurtre S, Martinot A, Duhamel A, Gauvin F, Grandbastien B,
Thi Vu N, et al. Development of a pediatric multiple organ dysfunction
score: use of two strategies. Med Decis Making 1999;19:399-410.
18. Leteurtre S, Martinot A, Duhamel A, Proulx F, Grandbastien B,
Cotting J, et al. Validation of the paediatric logistic organ dysfunction
(PELOD) score: prospective, observational, multicentre study. Lancet
2003;362:192-7.
19. Jones AE, Shapiro NI, Trzeciak S, Arnold RC, Claremont HA, Kline JA,
et al. Lactate clearance vs central venous oxygen saturation as goals of
early sepsis therapy: a randomized clinical trial. JAMA 2010;303:739-46.
20. Casserly B, Phillips GS, Schorr C, Dellinger RP, Townsend SR,
Osborn TM, et al. Lactate measurements in sepsis-induced tissue
hypoperfusion: results from the surviving sepsis campaign database.
Crit Care Med 2015;43:567-73.
21. Odom SR, Howell MD, Silva GS, Nielsen VM, Gupta A, Shapiro NI, et al.
Lactate clearance as a predictor of mortality in trauma patients. J Trauma
Acute Care Surg 2013;74:999-1004.
22. Murtuza B, Wall D, Reinhardt Z, Stickley J, Stumper O, Jones TJ, et al.
The importance of blood lactate clearance as a predictor of early
mortality following the modified Norwood procedure. Eur J
Cardiothorac Surg 2011;40:1207-14.
23. Munde A, Kumar N, Beri RS, Puliyel JM. Lactate clearance as a marker of
mortality in pediatric intensive care unit. Indian Pediatr 2014;51:565-7.
24. Ranjit S, Kissoon N. Bedside echocardiography is useful in assessing
children with fluid and inotrope resistant septic shock. Indian J Crit
Care Med 2013;17:224-30.
25. Scott HF, Donoghue AJ, Gaieski DF, Marchese RF, Mistry RD.
Effectiveness of physical exam signs for early detection of critical illness
in pediatric systemic inflammatory response syndrome. BMC Emerg
Med 2014;14:24.
26. Scott HF, Deakyne SJ, Woods JM, Bajaj L. The prevalence and diagnostic
utility of systemic inflammatory response syndrome vital signs in a
pediatric emergency department. Acad Emerg Med 2015;22:381-9.
27. Maitland K, Kiguli S, Opoka RO, Engoru C, Olupot-Olupot P,
Akech SO, et al. Mortality after fluid bolus in African children with
severe infection. N Engl J Med 2011;364:2483-95.
28. Ranjit S, Aram G, Kissoon N, Ali MK, Natraj R, Shresti S, et al.
Multimodal monitoring for hemodynamic categorization and
management of pediatric septic shock: a pilot observational study.
Pediatr Crit Care Med 2014;15:e17-26.
29. Sankar J, Sankar MJ, Suresh CP, Dubey NK, Singh A. Early goal-directed
therapy in pediatric septic shock: comparison of outcomes “with” and
“without” intermittent superior venacaval oxygen saturation
monitoring: a prospective cohort study. Pediatr Crit Care Med 2014;
15:e157-67.
30. Paul R, Neuman MI, Monuteaux MC, Melendez E. Adherence to
PALS sepsis guidelines and hospital length of stay. Pediatrics 2012;130:
e273-80.
31. Cruz AT, Perry AM, Williams EA, Graf JM, Wuestner ER, Patel B.
Implementation of goal-directed therapy for children with suspected
sepsis in the emergency department. Pediatrics 2011;127:e758-66.
32. Larsen GY, Mecham N, Greenberg R. An emergency department septic
shock protocol and care guideline for children initiated at triage.
Pediatrics 2011;127:e1585-92.
33. Brierley J, Carcillo JA, Choong K, Cornell T, Decaen A, Deymann A, et al.
Clinical practice parameters for hemodynamic support of pediatric and
neonatal septic shock: 2007 update from the American College of Critical
Care Medicine. Crit Care Med 2009;37:666-88.
34. Leteurtre S, Duhamel A, Salleron J, Grandbastien B, Lacroix J, Leclerc F,
et al. PELOD-2: an update of the PEdiatric logistic organ dysfunction
score. Crit Care Med 2013;41:1761-73.
35. Jones AE, Leonard MM, Hernandez-Nino J, Kline JA. Determination of
the effect of in vitro time, temperature, and tourniquet use on whole
blood venous point-of-care lactate concentrations. Acad Emerg Med
2007;14:587-91.
7
THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume -

Eligible for Screening

(n = 600)

ED care outside enrollment hours (54, 9.0%)

Screened
(n = 546, 91.0%)

Ineligible (436, 79.9%)

No organ dysfunction (402, 92.2%)
No suspicion for infection (28, 6.4%)
Inborn error of metabolism (6, 1.4%)

Eligible
(n = 110, 20.1%)

Missed Eligible (4, 4%)

Declined (21, 19%)

Enrolled
(n = 85, 77%)

Not Analyzed: Unable to Analyzed

collect lactate (n = 8, 9%) (n = 77, 91%)

Figure 1. Enrollment diagram.

7.e1 Scott et al
- 2015 ORIGINAL ARTICLES

Table I. Study organ dysfunction definitions, adapted from Goldstein et al12

Cardiovascular dysfunction (any of the following):
Despite isotonic intravenous bolus $40 mL/kg
Systolic blood pressure <5% for age or
Need for vasoactive drug (dopamine >5 mg/kg/min, or dobutamine, epinephrine, norepinephrine)
Capillary refill >5 s
Urine output <0.5 cc/kg/h
Respiratory dysfunction (any of the following):
PaO2/FiO2 <300 in absence of cyanotic heart disease or preexisting lung disease
PaCO2 >65 torr or 20 mm Hg over baseline
Proven need for >50% FiO2 to maintain saturation $92%
Neurologic dysfunction (any of the following):
Glasgow coma scale #11 or acute change $3 points below abnormal baseline
Hematologic dysfunction (any of the following):
Platelets <80 000 or decline of 50% from highest value over past 3 d in patients with baseline low platelets
International normalized ratio >2
Renal dysfunction:
Creatinine $2 times upper limit for age or 2-fold increase in baseline creatinine in patients with baseline elevations in creatinine
Hepatic dysfunction (any of the following):
Total bilirubin $4 (not applicable to newborn)
ALT 2 times upper limit of normal for age or 2-fold increase in baseline abnormal ALT

ALT, alanine transaminase; FiO2, fraction of inspired oxygen.

Table II. Selected characteristics of enrolled but not

analyzed subjects in whom 2- and 4-hour lactate levels
were unable to be collected
Characteristics Enrolled subjects not analyzed (8)
Baseline characteristics
Initial lactate level (mmol/L) 1.1 [1.1, 3.6]
Hypotension prior to enrollment 4 (50%)
Clinical outcomes
Hospital length of stay (d) 9.2 [4.0, 16.9]
PICU length of stay (d) 1.8 [0.5, 6.9]
In-hospital mortality 0 (0%)

PICU, pediatric ICU.

Continuous and ordinal variables reported as median [IQR]; categorical variables reported with
number (percentage).

Lactate Clearance and Normalization and Prolonged Organ Dysfunction in Pediatric Sepsis 7.e2
THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume -

Table IV. Subject characteristics, by lactate clearance or nonclearance status

Characteristics Lactate clearance (n = 70) Lactate nonclearance (n = 7) P value
Baseline characteristics
Age (y) 12.0 [6.0, 14.0] 7.0 [2.0, 13.0] .21
Chronic medical condition 47 (67%) 5 (71%) .99
Neuromuscular 25 (36%) 3 (43%)
Pulmonary 15 (21%) 3 (43%)
Endocrine/metabolic 8 (11%) 1 (14%)
Oncologic 8 (11%) 1 (14%)
Gastrointestinal/liver 9 (13%) 1 (14%)
Renal 4 (6%) 1 (14%)
Hematologic 3 (4%) 0 (0%)
Cardiac 2 (3%) 0 (0%)
Central line present 14 (20%) 1 (14%) .72
Initial lactate level (mmol/L) 2.1 [1.4, 3.0] 2.9 [1.6, 3.6] .30
Initial PELOD score 10.0 [10.0, 11.0] 12.0 [2.0, 20.0] .22
Number of ED organ system dysfunctions 1.0 [1.0, 2.0] 2.0 [1.0, 3.0] .02*
Hypotension prior to enrollment 40 (57%) 2 (29%) .15
Treatment characteristics
ED sepsis resuscitation activation 29 (41%) 3 (43%) .94
Time to antibiotics (min) 94.0 [53.0, 185.0] 89.0 [81.0, 157.0] .88
Total volume of isotonic fluid/patient weight delivered by 4 h (mL/kg) 41.5 [21.8, 55.2] 36.1 [23.1, 54.0] .87
Vasoactive agents between h 0 and 4 20 (29%) 1 (14%) .67
Clinical outcomes
Hospital length of stay (d) 5.6 [2.9, 10.8] 6.0 [5.7, 25.8] .22
PICU length of stay (d) 1.7 [0.0, 3.7] 2.7 [1.4, 21] .16
Positive-pressure ventilation d 0.0 [0.0, 1.8] 0.0 [0.0, 25.7] .32
Vasoactive agent d 0.0 [0.0, 1.0] 0.0 [0.0, 2.0] .96
Number of organ systems dysfunctional at 48 h 0.0 [0.0, 1.0] 1.0 [0.0, 3.0] .20
In-hospital mortality 1 (1%) 1 (14%) .12
Primary source of infection identified 48 (69%) 6 (86%) .67
Pneumonia 22 (44%) 2 (29%)
Urosepsis 7 (10%) 2 (29%)
Abdominal 9 (13%) 0 (0%)
Blood 10 (14%) 5 (71%)
Toxic shock 3 (4%) 0 (0%)
Skin/soft tissue/musculoskeletal 3 (4%) 0 (0%)
Meningitis 0 (0%) 1 (14%)
Microbiologic infection identified 38 (54%) 6 (86%) .23
Viral PCR positive 19 (27%) 1 (14%)
Bacteria culture positive 20 (29%) 4 (57%)
Fungal culture positive 2 (3%) 0 (0%)
Malaria positive 0 (0%) 1 (14%)

Continuous and ordinal variables reported as median [IQR], categorical variables reported with (percentage). P values calculated with Fisher exact statistic for categorical variables and Wilcoxon
rank-sum for continuous variables.
All microbiologic sources and chronic conditions in all subjects are listed, with some subjects having multiple sources and conditions.
*Indicates significance to the level of P < .05.

Table V. Initial lactate level obtained from central and

peripheral intravenous catheters
n Median IQR
All subjects
Central venous catheter 14 2.17 mmol/L 1.27-2.99 mmol/L
Peripheral intravenous catheter 63 2.14 mmol/L 1.57-3.18 mmol/L
Subjects with PELOD #10
Central venous line (PELOD #10) 5 2.33 mmol/L 1.24-2.90 mmol/L
Peripheral intravenous catheter 33 2.02 mmol/L 1.44-2.83 mmol/L
(PELOD #10)
Subjects with PELOD >10
Central venous line (PELOD >10) 9 2.00 mmol/L 1.19-3.18 mmol/L
Peripheral intravenous catheter 30 2.26 mmol/L 1.76-3.63 mmol/L
(PELOD >10)

All patients, and patients with PELOD score #10 and >10 are shown.

7.e3 Scott et al
- 2015 ORIGINAL ARTICLES

Table VI. Absolute and relative change in lactate level from the first to the final lactate level, by clearance and
normalization status
Lactate clearance Lactate nonclearance Lactate normalization Lactate non-normalization
(n = 70) (n = 7) (n = 62) (n = 15)
Absolute change in lactate level 0.73 mmol/L [ 0.3, 1.55] 0.55 mmol/L [0.47, 1.69] 0.58 mmol/L [ 0.27, 1.44] 0.33 mmol/L [ .19, .58]
% change in lactate level 32.1% [ 19.5, 55.2%] 35.3% [16.0, 61.9%] 33.4% [ 19.4, 58.1%] 11.9% [35.2%, 30.6%]

Results presented as median [IQR].

Lactate Clearance and Normalization and Prolonged Organ Dysfunction in Pediatric Sepsis 7.e4