Relationship Between Lactate Levels and Length of Hospital Stay in Infants with Lower Respiratory Tract Infection

Abstract

Background: Increased lactate concentrations are directly related to the severity of shock and mortality rates. There are limited data regarding the prognostic value of lactate among lower respiratory tract infections. We aimed to investigate the impact of lactate levels on admission on the clinical outcomes of children with lower respiratory tract infections.

Methods: We performed a retrospective study of hospitalized patients aged 1–12 months. We recorded data on patient demographics, clinical, laboratory, treatment, and outcomes. The primary outcome measure was the length of hospital stay, and the secondary outcome was transfer to the pediatric intensive care unit (PICU) and/or mortality rates.

Results: A total of 304 infants were included in the study. There were 198 infants with lactate levels of >2 mmol/L. Lactic acidosis was present in 6 infants, with a mean hospital stay of 8 ± 3 days. Only 1 (0.3%) patient required intubation, and 5 (1.6%) were transferred to the PICU. The overall mortality rate was 0%. Lactate levels (≤2 and >2 mmol/L) were not related to the length of hospital stay, transfer to PICU/discharge, and the need for intubation (P = 0.16, 0.8, and 0.46, respectively). The length of hospital stay was not correlated with lactate levels on admission (r = 0.01, P = 0.84), pCO₂ (r = 0.03, P = 0.52), pH (r = 0.07, P = 0.19), C-reactive protein (r = 0.06, P = 0.28), and oxygen saturation (r = −0.02, P = 0.72).

Conclusions: Lactate levels on admission did not predict the length of hospital stay in children with lower respiratory infections and were not related to the need for transfer to the intensive care unit. We suggest using lactate levels in combination with clinical, laboratory, and physical examination findings as predictors of disease severity.

Introduction

Blood lactate concentration has been widely used as a marker of altered tissue perfusion in states of shock. Lactate is produced daily from many organs such as the muscle, intestines, skin, red blood cells, and brain, and is mostly metabolized by the liver.^1,2 An increase in lactate levels has been shown to be directly related to the severity of shock and mortality rates.³

Lower respiratory infection is one of the leading causes of mortality in children younger than 5 years.⁴ The clinical severity of acute lower respiratory infection (ALRI) is most commonly evaluated by the presence of tachypnea, retractions, and low oxygen saturation (SO₂).⁵ Identification of predictors of morbidity and mortality is an important issue for these common infections. Studies have shown an association between serum lactate levels and mortality in critically ill children, mostly consisting of patients with cardiac anomalies, sepsis, shock, trauma, cardiac arrest, surgery, and those admitted in the pediatric intensive care unit (PICU).^2,3,6–10 There are limited data that analyzed the relationship between lactate levels and morbidity in the pediatric population with ALRI.

We aimed to investigate the relationship between serum lactate levels during admission and morbidity outcomes in pediatric patients with ALRI. We hypothesized that elevated lactate levels during admission could predict the length of hospital stay in children with ALRI.

Materials and Methods

This cross-sectional retrospective study included children from pediatric clinics between January 2019 and February 2020. Only hospitalized children diagnosed with severe ALRI, aged between 1 and 12 months, were included in this study. Newborns and outpatients with mild ALRI were excluded from the study. Indications for hospitalization in children with severe ALRI were as follows: toxic appearance, poor feeding, dehydration, respiratory distress, worsening or no response to outpatient therapy, underlying chronic conditions, and fever >38.5°C.¹¹ Presence of intercostal, subcostal, or suprasternal retractions, grunting, nasal flaring, SO₂ ≤92%, and respiratory rate >70 breaths/min were considered respiratory distress.¹¹ All of the hospitalized patients with clinical evidence of severe ALRI were routinely tested for blood gas analysis and chest X-ray.

Lactate measurements were performed immediately after the patient was admitted to the pediatric emergency department. Venous blood was collected using heparinized syringes. Normal lactate levels were <2 mmol/L based on our laboratory guidelines. Lactate levels were measured in venous blood using a blood gas analyzer (Radiometer ABL800; Copenhagen, Denmark), which displayed lactate ranges of 0–30 mmol/L. Lactic acidosis was defined as lactate levels of >5 mmol/L and pH of <7.35.¹² Any known chronic medical illnesses (inborn errors of metabolism, liver dysfunction, and mitochondrial cytopathy) that can affect lactate levels were not included in the study.

A review of data on the demographics, clinical, laboratory, treatment, and outcomes was recorded. Clinical data were extracted from medical records, which included age, sex, SO₂, presence of comorbidities, laboratory data [including white blood count, neutrophil counts, C-reactive protein (CRP), venous blood gases (pH, pCO₂, and lactate), and documented microorganisms by respiratory polymerase chain reaction (PCR) results], treatment modalities [antibiotics, nebulized bronchodilators (salbutamol, corticosteroids, and ipratropium bromide), and oxygen supplementation], PICU admission, and outcome data about mortality and length of admission. In our institution, the need for transfer to the PICU, criteria for hospitalization, and suitability for discharge were determined based on the community-acquired pneumonia guidelines developed by the Turkish Thoracic Society.¹¹

This study was approved by the hospital's ethics committee Approval no:170.

Statistical analyses

Statistical analyses were performed using SPSS version 17 (IBM SPSS Statistics, Chicago, IL). Data were expressed as numbers, percentages, and mean ± standard deviation, as needed. The Kolmogorov–Smirnov test was used for the normality analysis. Independent-group t-tests were used to compare continuous variables between the groups. A binary comparison between children with normal and hyperlactatemia (lactate levels ≤2 versus >2 mmol/L) was performed using the χ² test for categorical variables. Correlation analysis for continuous variables (eg, hospital length of stay versus lactate levels) was conducted using Pearson's s analyses. Statistical significance was set at P < 0.05. The primary outcome measure was the length of hospital stay, and secondary outcomes were the transfer to PICU and/or mortality.

Results

There were 864 children hospitalized during the study period, 35% of whom were younger than 1-year old. A total of 304 infants were included in this retrospective study. The median age was 4 months (range: 1–12 months), and 195 (64.1%) were male. The clinical and laboratory characteristics of the children are shown in Table 1.

Table 1.

Clinical and Laboratory Findings of the Patients

Characteristics	Mean ± SD	Minimum	Maximum
Age in months	4.3 ± 2.8	1	12
Length of hospital stay, days	7.4 ± 3.3	1	29
Use of antibiotic, days (n = 293)	5.1 ± 3.8	0	27
Use of parenteral fluid, days (n = 293)	3.3 ± 2.5	0	28
Use of steroid, prednisolone i.v., days (n = 171)	1.6 ± 1.7	0	7
Use of magnesium i.v., days (n = 19)	0.1 ± 0.3	0	3
Saturation at admission, %	95.6 ± 3.3	73	100
Use of bronchodilator support, days
Salbutamol (n = 282)	5.9 ± 2,9	0	22
Ipratropium bromide (n = 26)	0.4 ± 1.7	0	18
Steroid, budesonide (n = 232)	5.2 ± 3.9	0	27
Hypertonic saline, 3% NaCl (n = 20)	0.3 ± 1.3	0	8
Laboratory findings
WBC,/mm3	11,481 ± 4,976	1,750	36,280
Neutrophil %	41.8 ± 18.2	5	90
pH	7.38 ± 0.04	7.22	7.55
CRP, mg/dL	1 ± 2.3	<0.02	34
pCO2, mm Hg	36.6 ± 6.5	16	76
Serum lactate, mmol/L	2.6 ± 1.1	0.8	7.7

CRP, C-reactive protein; SD, standard deviation; WBC, white blood count.

The length of hospital stay was not correlated with lactate levels during admission (r = 0.01, P = 0.84), pCO₂ (r = 0.03, P = 0.52), pH (r = 0.07, P = 0.19), CRP (r = 0.06, P = 0.28), and SO₂ (r = −0.02, P = 0.72).

Clinical and laboratory data were compared based on lactate groups (lactate values ≤2 and >2 mmol/L) (Tables 2–3). Only 1 (0.3%) patient required intubation, and 5 (1.6%) were transferred to the PICU. Lactic acidosis was present in 6 infants, with a mean hospital stay of 8 ± 3 days. All patients with lactic acidosis were discharged from the hospital, and none of them was transferred to the PICU. The mortality rate was 0%.

Table 2.

Comparison of the Categorical Characteristics and Outcomes of Children Based on Lactate Levels (n = 304)

	Lactate values		P–
	≤2 mmol/L	>2 mmol/L
Outcome of the patients
Transfer to PICU (n = 5)	2	3	0.8
Discharge (n = 299)	104	195
Intubation
Yes (n = 1)	0	1	0.46
No	106	197
Comorbidities
Yes (n = 59)	21	38	0.89
No	85	160

PICU, pediatric intensive care unit.

Table 3.

A Binary Comparison of Continuous Variables Between Children with Normal Lactate Levels and Hyperlactatemia

	Lactate values		P–	95% confidence interval
	≤2 mmol/L	>2 mmol/L		Lower	Upper
Age, months	4.6 ± 3	4.1 ± 2.7	0.13	−0.1	1.2
Length of hospital stay, days	7.8 ± 3.7	7.2 ± 3.1	0.16	−0.2	1.3
Use of antibiotic, days	5.2 ± 4.1	5 ± 3.7	0.74	−0.7	1
pH	7.4 ± 0.03	7.37 ± 0.04	0.000	0.02	0.04
pCO2, mm Hg	36 ± 7	37 ± 6	0.22	−2.5	0.5
Neutrophils, %	41 ± 16	42 ± 19	0.53	−5.4	2.8
CRP (mg/dL)	0.8 ± 1	1.2 ± 2.8	0.09	−0.8	0.06
Saturation on admission, %	95 ± 3	95 ± 3	0.68	−1	0.6

Bold value indicates statistical significant, P-value under 0.05.

CRP, C-reactive protein.

Analysis of PCR results of respiratory specimens was available for 210 (69.1%) children. Pathogens were detected in 163 samples (77.6%), and 47 (22.3%) revealed no pathogens. The most common viruses detected from the respiratory specimens were as follows: 23.8% RSV, 23.8% rhinovirus, 10.9% parainfluenza, 4.7% bocavirus, and 4.2% metapneumovirus. The type of microorganism was not related to either the length of hospital stay or blood lactate levels (P = 0.14 and 0.66, respectively).

Comorbidities were present in 59 (19.4%) children, including 12 (3.9%) with trisomy 21, 5 (1.6%) with congenital heart disease, and 25 (8.2%) with problems associated with prematurity. Patients with comorbidities had a significantly longer length of hospital stay than patients with no underlying disease, 9 ± 5.4 days versus 7 ± 2.5, P = 0.008 [95% confidence interval (CI), −3.4 to −0.5]. The presence of comorbidities was not related to lactate levels during admission (P = 0.55; 95% CI, −0.4 to 0.22). The presence of comorbidities was related to the need for transfer to the PICU (P = 0.02) and use of oxygen supplementation (P = 0.00).

Discussion

In this study, we aimed to evaluate the relationship between lactate levels on admission and the outcome of pediatric patients with ALRI. We noted that lactate levels during admission were not correlated to the length of hospital stay or the need for transfer to the PICU.

In clinical practice, serum lactate levels have been routinely used as a marker of tissue perfusion and guide therapeutic decisions in critically ill patients.^1,2 Persistent elevation of lactate levels in both adult and pediatric patients is associated with a high patient mortality rate and multiple organ damage. Existing literature mostly focuses on the relationship between lactate levels and mortality risk among critically ill patients, rather than morbidity.^3,6–9

Studies on the role of lactate with ALRI are sparse. There are 2 studies from underdeveloped countries, Uganda and Malawi.^5,13 A prospective cohort study from Uganda included 155 hospitalized children with clinical signs of pneumonia.¹³ Lactate levels were measured at admission and daily using a handheld device from a finger prick blood test during hospitalization. Five days after hospitalization, the mortality rates were 2%, 11%, and 26% among children with admission lactate levels of <2.0, 2.0–4.0, and >4.0 mmol/L, respectively.¹³ The in-hospital mortality rate was 14%. This high mortality rate could be attributable to subgroups of patients, including those with respiratory distress and HIV seropositivity with pneumonia. The authors concluded that a rapid lactate measurement was an accurate predictor of mortality among children with pneumonia admitted to hospitals and suggested integration of lactate measurements in the pneumonia management algorithms in low-income countries.¹³

Ramakrishna et al.⁵ evaluated the predictive value of lactate levels on admission on the mortality risk of children with pneumonia. The median serum lactate concentration was 2.7 mmol/L (interquartile range 1.8–4.4 mmol/L) among 233 children with pneumonia.⁵ There were 10 deaths out of 77 (13%) children who had a serum lactate concentration of 2.1–4.0 mmol/Land 13 deaths out of 72 (18%) children who had lactate levels of >4.0 mmol/L. When the lactate level was above 2 mmol/L, the relative risk of death was 7.48 (1.72–32.6).⁵ Children with hypoxemia and prolonged capillary refill time were predisposed to have hyperlactatemia. Hyperlactatemia was found to be a risk factor for death in children with severe pneumonia in Malawi.⁵

In this study, we included only infants who necessitated hospitalization for ALRI. We observed no mortality and only 5 infants were transferred to the PICU. The median lactate level was 2.5 mmol/L, similar to the results of Ramakrishna et al.⁵ The presence of hypoxemia and prolonged capillary refill time in hyperlactatemic children may be the reason for the higher mortality rates seen in previous studies compared with our results. In addition, early and easy access to health care facilities in our country and the availability of PICUs may lower mortality rates.

There is no consensus about a standard threshold lactate level for pediatric patients. In the year 2020, the optimal threshold to define “hyperlactatemia” remains unclear.¹⁴ Most literature has studied the association of lactate levels in patients with sepsis, cardiac conditions, shock, trauma, and surgery in the PICU or general ICUs.^2,6–10 Lactate levels should be evaluated as part of a more comprehensive assessment to determine a patient's clinical and perfusion status.¹⁴ Physical examination signs, SO₂, and lactate levels could play an important role in determining patient prognosis with pneumonia in developing countries.⁵

Our study has some limitations. We did not study serial blood lactate concentrations because of the retrospective study design; thus, collecting only the initial lactate levels may not represent the patient's outcome and cannot reflect the changes over time of lactate levels.

Despite the fact that most of the children included in this cohort had a final diagnosis of bronchiolitis and viral pneumonia, the use of antibiotics, systemic/inhaled steroids, and inhaled bronchodilators was very high. Although bronchiolitis guidelines advise against the routine use of medicines,¹⁵ it is still often prescribed. Age, preterm infants, chronic lung disease of prematurity, congenital heart disease are important predictors of disease severity.¹⁵ While in our cohort, empiric antibiotics, bronchodilators, and steroids might be indicated in children with comorbidity (19.4% of our cohort), we need to review our current practice.

This was a single-center study, which limits the generalizability of our results. In addition, we did not investigate the outcomes of children transferred to PICUs in other institutions, which may change the mortality rate. The lack of a relationship between lactate levels and the length of hospital stay may be due to the low rate of children with lactic acidosis on admission. At the time of writing, to our knowledge, there are no published data on the relationship between admission serum lactate levels and the length of hospital stay in children diagnosed with ALRI.

Conclusions

Lactate is an accurate, fast, easy, and cost-effective predictor of patient outcomes. Currently, there are no randomized-controlled studies that have evaluated whether initial or serial measurement of blood lactate levels better guides evaluation and management in children. Until more definitive studies become available, we suggest using lactate levels in combination with clinical, laboratory, and physical examination findings to guide the management decisions of patients with ALRI.

Authors' Contributions

F.E., S.S., E.T., and E.U.Y. contributed to the conception and design, and acquisition of data; F.E., S.S., and E.T. analyzed chest X-rays; E.U.Y. and R.G.S.Y. contributed to the interpretation of data. R.G.S.Y. and E.U.Y. wrote the first draft of the article. All of us have participated in critical revision of the article for important intellectual content and final approval of the version to be published.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.