Journal Logo

Feature Articles

Impact of Blood Sampling on Anemia in the PICU: A Prospective Cohort Study

François, Tine MD1; Sauthier, Michaël MD, MBI1; Charlier, Julien MD1; Dessureault, Jessica BSN, RN1; Tucci, Marisa MD, FRCPC, FAAP1; Harrington, Karen MD, MSc1; Ducharme-Crevier, Laurence MD, FRCPC, MSc1; Al Omar, Sally PhD2; Lacroix, Jacques MD, FRCPC, FAAP1; Du Pont-Thibodeau, Geneviève MD, FRCPC, MSc1

Author Information
Pediatric Critical Care Medicine: June 2022 - Volume 23 - Issue 6 - p 435-443
doi: 10.1097/PCC.0000000000002947



  • More than 50% of PICU survivors are anemic at discharge. Anemia has well-known negative impacts on neurocognitive development and quality of life.
  • Etiology of anemia in critically ill patients is multifactorial. Diagnostic blood sampling is a modifiable contributor.
  • Diagnostic blood sampling and wasted volume during sampling are significant in PICU and associated with the prevalence of anemia at discharge. Cardiac surgery patients and children with sepsis/shock are more at risk of high blood sampling volumes.

More than 50% of PICU survivors are anemic at discharge (1–4). It is unknown whether acute anemia developed in the ICU is dangerous for children in the long term. The trajectory of anemia in PICU survivors is also unknown. A large proportion of ICU adults who are anemic at discharge remain anemic for many months (5,6). Given that anemia can have a negative impact on neurocognitive development and quality of life in children, a detailed evaluation of potentially modifiable contributors to post-PICU anemia is essential (7–11).

Contributors to critical illness-associated anemia are numerous and include, among others, diagnostic blood sampling (12,13). This is well recognized as a cause of anemia in critically ill adults (14,15); however, data in children are scarce. Although strategies to reduce iatrogenic blood losses exist in the literature, additional data are required to better characterize the practice patterns of pediatric intensivists before targeted interventions can be introduced into their clinical practice.

This study aimed to quantify the blood volume removed for diagnostic testing (including volume discarded) in PICU children and to identify patient characteristics associated with higher blood removal. We also aimed to evaluate the possible association between blood sampling and the rate of anemia at PICU discharge as well as with a change in hemoglobin (Hb) from PICU entry to discharge.

We hypothesized that the amount of blood volume sampled and discarded would be significant and that this would be associated with anemia at PICU discharge.


We conducted a prospective observational cohort study in critically ill children at the PICU of Sainte-Justine University Health Centre, Montreal, QC, Canada.

All consecutive admissions from September 13, 2019, to January 15, 2020, were considered eligible for the study, using following exclusion criteria: premature neonates (gestational age < 37 wk at PICU admission), adult patients (> 18 yr old), and pregnant or immediate postpartum patients. Patients with known anemia or coagulation disorders were not excluded. Readmissions to PICU within 48 hours were considered as one admission; patients readmitted after 48 hours were considered as a new PICU admission.

The study protocol was approved by the Institutional Review Board of the Sainte-Justine University Hospital Research center (2020-2380), which waived the need for informed consent.

Data Collection and Management

For 4 months, blood volume sampled for diagnostic laboratory analysis, including discarded blood not sent to the laboratory for assay, was prospectively recorded in the electronic medical record (IntelliSpace Critical Care and Anesthesia, F.01, Philips, Amsterdam, The Netherlands) by the nursing team for the entire PICU stay of each patient. Blood volumes withdrawn but not registered by the nurse were estimated by calculating the minimum standard volume required for each laboratory test performed. In our unit, blood sampling from an arterial line is done by a closed in-line sampling system (TruWave, Edwards Lifesciences, Germany), but the sampling point needs to be cleared (NanoClave Luer Lock, ICU Medical, San Clemente, CA). Venous line sampling is performed using an open sampling technique. The volume of blood systematically discarded to ensure reliable blood sampling is: 1 mL for central venous line sampling, 0.1 mL for arterial lines, 0.05 mL for capillary sampling, and 0 mL for venous/arterial punctures. Discarded blood volumes not registered by the nurse were estimated by counting the number of blood draws and the specific blood volumes discarded according to sample site. A post hoc Bland-Altman analysis showed a strong and statistically significant correlation between the measured (reference) and estimated volumes of blood drawn (r = 0.85, p < 0.0001) and discarded blood (r = 0.71, p < 0.0001). Total circulating blood volume (TCV) was based on weight and age: 80 mL/kg for children less than 2 years and 70 mL/kg for patients aged 2-18 years (16,17).

Data collection further included: patient demographics, admission diagnosis, comorbidities, admission and worst daily Pediatric Logistic Organ Dysfunction-2 score (PELOD-2) (18), PICU length of stay (LOS), ventilatory support, surgical interventions during stay, arterial or central venous access, Hb level at admission and prior to discharge, and RBC transfusion.

Admission diagnoses were recorded as documented in the patient chart and categorized as follows: cardiac surgery, noncardiac surgery, respiratory, neurologic, sepsis (including severe infection and systemic inflammatory response syndrome), and shock. Other diagnoses with need for cardiorespiratory monitoring or specific treatment (e.g., intoxication, trauma, diabetic ketoacidosis, and electrolyte disturbances) were recorded as “other.”

Our unit follows recommended restrictive transfusion guidelines (19); we generally avoid RBC transfusion for hemodynamically stable, noncyanotic patients with an Hb above 70 g/L, although the decision remains at the discretion of the attending physician.

Outcome Measures

Sampled blood volume was reported in milliliters per kilogram bodyweight per patient per day as well as per patient stay. This included blood used for laboratory measurements as well as discarded blood. We recorded the number of blood draws per patient per day and per PICU stay.

We investigated the association of blood sampling with anemia incidence at PICU discharge. Anemia definitions were based on age following the Canadian Blood Services Criteria (20) (Supplemental Table 1, Presence or absence of anemia at PICU discharge was based on the last Hb level measured prior to PICU discharge. Based on clinical experience and existing pediatric literature (1–3), the following additional risk factors or possible confounders for anemia at discharge were considered: age, sex, worst daily PELOD-2, LOS, admission diagnosis, anemia at PICU entry, and transfusion.

To eliminate bias caused by the presence of anemia at admission and by the administration of transfusions, we also investigated the impact of blood sampling on the absolute change in Hb level from PICU admission to discharge using the first and last Hb levels during PICU stay.

Statistical Analysis

Mean values (± sd), medians (interquartile range [IQR]), and proportions (%) were used for descriptive analyses. Wilcoxon two-sample tests were used to identify any unadjusted association between the categorical variables and outcome parameters as well as any correlation with proportional blood sampling volumes. Any monotonic correlation between continuous variables and predefined outcomes was identified by Spearman correlation test. Univariate and multivariate logistic regression analyses were used to assess the association with anemia at discharge and the association with absolute Hb change from PICU admission to discharge. Patients with only one or no Hb measurement during PICU were excluded from all analyses pertaining to investigate the association of blood sampling with anemia or Hb change. A priori selected potential determinants were added to the regression model. Statistical power was computed preanalysis at 80% to detect an odds ratio greater than or equal to 2.2 on a sample size of 320 patients. Two-sided thresholds for significance (p) were preset at 0.05 for all tests. All statistical analyses were performed using the Statistical Analysis System statistical software (Version 9.4; SAS Institute, Cary, NC).


From September 2019 to January 2020, there were 435 consecutive PICU admissions, including seven readmissions within 48 hours of PICU discharge. Five patients were excluded due to age and obstetrical admission diagnosis; 423 admissions were retained, of which 314 had enough data to be included in the multivariate analysis (Fig. 1).

Figure 1.:
Patient enrollment.

Patient Demographics

Patient characteristics at PICU entry and hospitalization are reported in Table 1. Respiratory failure was the most common medical admission diagnosis, whereas one-third of patients were admitted postoperatively. Sixteen patients underwent surgery in PICU. Four patients died before discharge.

TABLE 1. - Patient Characteristicsa
Variable All Patients (Enrolled at PICU Entry) (n = 423) Anemia at PICU Discharge (n = 177) No Anemia at PICU Discharge (n = 138) < 2 Blood Samples(During PICU Stay) (n = 102)
At PICU entry
 Age, mo 59.1 (±66.9); 28.0 (4.0–108.0) 74.3 (±67.7); 54.0 (12.0–133.0) 41.5 (±62.6); 9.5 (2.0–56.0) 28.8 (±69.0); 27.0 (8.0–88.5)
 Weight, kg 20.3 (±20.7); 12.7 (6.2–24.6) 24.9 (±23.0); 15.9 (8.6–31.0) 15.1 (±18.2);7.2 (4.0–16.7) 19.8 (±19.5); 12.5 (7.6–21.0)
 Sex (male), n (%) 229 (54) 100 (58) 72 (42) 55 (54)
 Hb at PICU entry,b g/L 117 (±23); 114 (103–130) 106 (±18); 108 (96–116) 131 (±25); 129 (113–146) -
 Anemia at PICU entry,bn (%) 161 (41)b 121 (86) 19 (14) -
 Main admission diagnosis
  Neurologic disease 35 (8) 16 (9) 12 (9) 4 (4)
  Other 59 (14) 24 (14) 12 (9) 19 (19)
  Cardiac surgery 50 (12) 29 (16) 19 (14) 0 (0)
  Noncardiac surgery 90 (21) 41 (23) 13 (9) 27 (26)
  Sepsis and shock 33 (8) 33 (8) 12 (9) 4 (4)
  Respiratory disease 156 (37) 49 (28) 69 (50) 48 (47)
During PICU stay
 Pediatric Logistic Organ Dysfunction-2 scorec 4.9 (±3.9); 4.0 (2.0–7.0) 4.8 ± 3.8; 4.0 (2.0–7.0) 5.5 (±4.3); 4.0 (2.0–8.0) 4.7 (±3.8); 3.0 (2.0–6.8)
 Length of stay, d 4.5 (±10.0); 2.1 (1.1–3.7) 4.9 ± 11.4; 2.2 (1.5–3.5) 5.3 (±10.5); 3.0 (1.4–5.0) 1.8 (±3.0); 1.1 (0.8–1.9)
 Invasive ventilation,dn (%) 126 (30) 76 (43) 41 (30) 9 (9)
 Noninvasive ventilation,dn (%) 198 (47) 74 (42) 84 (61) 42 (41)
 Vascular catheters,en (%) 179 (42) 103 (58) 54 (39) 14 (14)
  Central venous catheter, n (%) 134 (32) 73 (41) 50 (36) 11 (11)
  Arterial line, n (%) 120 (28) 81 (46) 31 (23) 3 (3)
 Blood samples per day, n 2.9 (±3.4); 2.0 (0.9–3.7) 3.7 (±3.0); 3.1 (1.8–4.7) 2.9 (±3.7); 1.8 (1.0–3.4) -
 Blood sample volume per stay, mL/kg 3.9 (±19.0); 0.4 (0.1–1.3) 5.8 (±27.3); 0.7 (0.3–1.7) 3.6 (±9.7); 0.6 (0.3–2.4) -
Transfusion, n (%) 43 (10) 27 (15) 15 (11) 2 (2)
 Hb at PICU discharge,f g/L 110 (±20); 109 (98–122) 98 ± 13; 99 (89–108) 126 (±17); 123 (113–134) -
 Anemia at discharge,fn (%) 177 (56)f 177 (100) 0 (0) -
Hb = hemoglobin.
aData are reported in mean (± sd) and median (interquartile range) for continuous data; or number (%) for dichotomous data.
bTotal number of patients with hemoglobin at PICU entry: n = 392.
cPediatric Logistic Organ Dysfunction score; worst daily score during hospitalization.
dInvasive ventilation: intubated or tracheotomized patient supported by mechanical ventilation; noninvasive ventilation: support by high-flow nasal cannula or any positive-pressure support.
eVascular catheters: all central venous catheters or peripherally inserted central catheters, central or peripheral arterial catheters.
fTotal number of patients with hemoglobin at PICU discharge: n = 316.

Median LOS was 2.1 days (IQR, 1.1–3.7 d) with an average LOS of 4.5 (sd ± 10.0) days. Invasive ventilation was used in 30% of patients and noninvasive ventilatory support in 50%. Nearly 40% of patients had a central venous or arterial line, with a mean duration of catheter of 8.5 (± 14.4) and 3.7 (± 6.8) days, respectively.

Blood Sampling Practices

Overall, patients had two blood collections per day (IQR, 0.9–3.7). Almost 10% of patients had no blood sampled, and 14% had only one blood sample. Blood volumes were recorded for 57% of all blood samples.

Mean blood volume sampled was 5.3 (± 6.3) mL/patient/d (median, 3.0 mL [21]). Blood volume sampled per PICU stay was on average 3.9 (± 19.0) mL/kg (median, 0.4 mL/kg [22]), which corresponds to approximately 5% of TCV. In 30 of 423 patients (7.1%), total blood volume sampled was greater than 10% of TCV. On an average, 1.0 (± 5.9) mL/kg of blood was discarded per PICU hospitalization (median 0.05 mL/kg [23]), which represents 26% of total blood sampled.

Patient Characteristics Associated With Blood Sampling

Patients admitted following cardiac surgery and those with sepsis or shock had the largest blood volume sampled (11.0 ± 4.5 and 8.7 ± 8.8 mL/d, respectively; p < 0.001). Patients with a primary respiratory diagnosis had the least blood sampled (2.1 ± 3.4 mL/patient/d). Patients with ventilatory support had significantly more blood sampled than those without (38.0 ± 116.5 mL vs 10.5 ± 16.9 mL/admission; p < 0.001). Total sampled blood volume was significantly greater in patients with a vascular access than in patients without (8.2 ± 6.4 vs 3.1 ± 5.3 mL/patient/d; p < 0.001). Patients with a central venous line had a significantly greater volume of blood discarded (14.4 ± 35 vs 2.1 ± 6.9 mL/admission; p < 0.001). We were unable to detect an association between increased blood sampling and severity of illness (worst daily PELOD-2) (R = –0.044; p = 0.43).

Impact of Blood Sampling on Anemia at PICU Discharge

Forty-three patients (10%) received at least one RBC transfusion during PICU stay, with the highest number of transfusions reported in septic and cardiac surgery patients. Most children received their transfusion within the first 24–48 hours (median, 11.8 hr; IQR, 5–41 hr) after PICU admission. More than half of patients were anemic at PICU discharge (177/315, 56%), with 16% being severely anemic (Hb < 90 g/L). Eighteen percent (57/314) had new onset of anemia at PICU discharge.

Risk factors associated with anemia at discharge were identified using a multivariate logistic regression model (Supplemental Table 2, We detected a significant association between the volume of blood sampled (as a continuous logarithmically transformed variable) and anemia at discharge (adjusted odds ratio [aOR], 1.63; 95% CI, 1.18–2.45; p = 0.003). When categorizing blood sample volumes, we observed a trend toward higher risk of anemia at discharge in patients sampled 1 to 5 mL/kg and greater than 5 mL/kg versus those sampled less than 1 mL/kg (aOR, 2.22; 95% CI, 0.96–5.12 and aOR, 3.10; 95% CI, 0.88–10.94, respectively). Children with anemia at PICU entry were significantly more at risk for anemia at discharge (aOR, 13.52; 95% CI, 6.78–26.97). We observed a significant association between age and anemia at discharge, with adolescents (>10 yr old) being at greater risk (aOR, 6.53; 95% CI, 2.02–21.04).

Absolute Hemoglobin Change From PICU Entry to PICU Discharge

Risk factors associated with an absolute change in Hb (ΔHb) during PICU were identified using a multivariate linear regression model (Supplemental Table 3, Children without anemia at PICU entry and untransfused patients were more at risk of having an Hb decline during PICU stay: 1) no anemia at admission: mean ΔHb –9.73 g/L (95% CI, –13.5 to –5.9 g/L) and 2) no transfusion: ΔHb –10.5 g/L (95% CI, –13.5 to –7.7 g/L). Children aged greater than 120 months were also more likely to experience an Hb decline (ΔHb, –6.4 g/L; 95% CI, –11.1 to –1.7). We did not find a statistically significant association between blood volume and Hb decline. Children sampled for greater than 5 mL/kg had an Hb change of –8.1 g/L (95% CI, –13.9 to –2.3) versus –2.6 g/L (95% CI, –6.4 to –1.2 g/L) for those sampled less than 1 mL/kg (p = 0.32). Supplemental Table 4 ( presents the PICU admission to PICU discharge change in Hb in patients depending on Hb status “throughout” PICU stay and those that: 1) developed a new anemia at PICU discharge, 2) had no change in anemia status during PICU stay, and 3) had anemia at admission but not at discharge. Children that developed a new anemia at PICU discharge had a greater drop in Hb from PICU admission to PICU discharge (p < 0.001). Supplemental Table 5 ( shows the anemia status of PICU children at PICU admission and at PICU discharge in proportions.

  • Diagnostic blood testing can account for nearly 5% of total circulating blood volume in children. A significant part of blood sampled is wasted at the bedside, mostly related to the sampling technique.
  • Blood volume sampled is significantly associated with a higher risk for anemia at PICU discharge, especially when sampling greater than 1 ml/kg.
  • Daily evaluation of the necessity of diagnostical blood testing and improvement of clinical practice and blood sampling techniques are necessary when caring for critically ill children.


Our study demonstrates that in PICU children, nearly 5% of TCV is removed for diagnostic testing and approximately 25% of blood drawn is discarded. Cardiac surgery patients and those in sepsis or shock have the greatest blood volumes sampled. Children with vascular access or ventilatory support have increased sample volumes. Blood sampling is associated with anemia at PICU discharge.

Diagnostic testing in critically ill patients helps monitor life-threatening and rapidly evolving illness and guides therapeutic interventions. Although blood sampling may be unavoidable, it may be possible to safely decrease the volume of blood sampled and potentially reduce the risks for anemia. Strategies include reducing blood sample frequency and the volume required for each diagnostic test, and minimizing blood waste during sampling (24). In our population, 25% of blood sampled was discarded at the bedside to obtain undiluted blood from central lines. This study is the largest cohort study to report on blood wasting and the only study to collect data on the blood wasted at the bedside (in addition to the blood wasted in the laboratory) (25–27). The proportion of blood wasted that we measured is even higher than previously described (25,26). This suggests that minimizing blood wasting should be an important intervention to introduce into clinical practice. Blood conservation devices can help to reduce the amount of blood wasted for line rinsing (21,28–33). Arterial closed blood sampling devices such as the Venous-Arterial blood Management Protection device can significantly lower the amount of blood volume discarded (28–30,34). Returning hemodiluted discarded blood with the ErythroSave device (a disposable sterile syringe that avoids the blood to cloth) has also been shown to significantly lower transfusion requirements in neonates (33). Small volume tubes, as used in our unit, and bedside point-of-care tests may reduce the blood volume required for testing, especially for coagulation testing that generally requires more blood volume than other tests (35). Other strategies include physician education and practice modification including elimination of routine blood tests, and daily assessment of blood testing needs during patient rounds (22,23,36). Noninvasive monitoring of physiologic markers (end-tidal CO2 and near-infrared spectroscopy) may also help reduce blood testing (37–40). In our study, patients with central line access were sampled the most; however, we lacked power to determine whether this was associated with severity of illness or simply because of convenience. Adult studies suggest central line access in itself regardless of severity of illness that may contribute to increased blood drawing (41). Early removal of central lines is part of physician education strategies to reduce blood testing (23,36).

Interventions should specifically focus on children undergoing cardiac surgery and those with sepsis or shock. Neonates undergoing cardiac surgery are particularly at risk of the physiologic complications of anemia given their underlying primary disease. Studies on children admitted following cardiac surgery have shown that bundles of interventions including education and practice changes reduce the number of blood tests and transfusion requirements (42–45). In septic patients, lactate and mixed venous oxygen saturation monitoring are used to monitor patient progress but significantly augment the amount of blood discarded when drawn from central venous lines. Safely reducing the frequency and the volume of blood loss in these higher risk populations and in all critically ill children is a recommendation of Choosing Wisely, a Canadian initiative exploring the diagnostic overuse phenomenon with up-to-date recommendations (46).

Blood testing was associated with the incidence of anemia at PICU discharge. Our multivariate analysis was controlled for the presence of anemia at admission and for transfusions to avoid interaction with potential confounders. We additionally investigated the impact of blood sampling on Hb change from PICU admission to discharge, to eliminate the inevitable interaction between the presence of anemia at admission and the incidence of anemia at PICU discharge. We unfortunately lacked power to detect an association between blood sampling and Hb change despite a clear trend (Supplemental Table 3, A nonsignificant association does not allow us to conclude in the absence of association. However, we did find that children that develop a new anemia at PICU discharge had a significant drop in their Hb from PICU admission to PICU discharge (Supplemental Table 4, This suggests that the presence of anemia at PICU admission alone is certainly not the only risk factor for having anemia at PICU discharge. Anemia is an outcome of concern for many reasons. Anemia, especially iron-deficiency anemia, is associated with irreversible adverse effects on the neurocognitive development of young children (9–11). It is also associated with a poorer quality of life, fatigue, muscular weakness, and exercise intolerance (8). Preventative strategies such as IV iron or erythropoietin have been investigated to prevent the development anemia while patients are in the ICU, but results were not significant (47–49). Anemia at PICU discharge is multifactorial and can be partly caused by nonmodifiable etiologies such as inflammation from critical illness (50); however, reducing diagnostic blood sampling and waste is a certainly modifiable risk factor and one that can be addressed.

This study has several strengths. First, it is the largest prospective study on the matter. Data on iatrogenic anemia in general PICU children are scarce; we found only four studies published on this topic (1–3,51). Second, ours is the first to report the proportion of discarded blood at the bedside in a large cohort of consecutively admitted PICU patients. Third, in contrast to previous studies, we were able to detect a significant association between the blood volume sampled and anemia at PICU discharge (1).

This study does have some limitations. First, to preserve the observational nature of the study and to avoid additional blood work, we did not request any supplemental blood tests. Consequently, Hb values close to PICU discharge were lacking in 25% of patients. Despite this, we observed a similar proportion of anemic patients at PICU discharge as reported in the literature (1–3). Second, nursing and medical personnel were aware of the study, and bedside nurses recorded blood volume sampled in the medical record. It is conceivable that this affected the way the team ordered blood work, sampled patients, or reported their practices. If so, the result would be an underestimation of actual blood sampling practice. Third, the statistical results using logarithmic transformed continuous blood sample data are difficult to translate into clinical significance. Therefore, we chose to categorize blood sample data to facilitate clinical interpretation, although this led to a loss in statistical power. Fourth, we did not investigate further the relationship between blood sampling and transfusion. The overall proportion of patients transfused was low, and most received their transfusion within 24–48 hours after PICU admission. Finally, even though mechanical ventilation is a known surrogate for severity of illness and ventilated children in our cohort were sampled more, most children had low PELOD scores of 1–4 (Supplemental Table 2,; and Supplemental Table 3, This made our study insufficiently powered to detect a significant association of severity of illness on anemia.


Diagnostic blood volumes sampled and blood volumes discarded are significant in critically ill children and are associated with anemia at PICU discharge. Studies are necessary to determine which interventions can reduce diagnostic testing safely and rapidly in critically ill children to avoid unnecessary complications.


We thank the nursing team of our PICU for their essential contribution.


1. Bateman ST, Lacroix J, Bowen K, et al.; Pediatric Acute Lung Injury and Sepsis Investigators Network: Anemia, blood loss, and blood transfusions in North American children in the intensive care unit. Am J Respire Crib Care Med 2008; 178:26–33
2. Demarest P, Karma O, Tucci M, et al.: Anemia at pediatric intensive care unit discharge: Prevalence and risk markers. Ann Intensive Care 2017; 7:107
3. Demarest P, Valla FV, Behal H, et al.: Anemia at discharge from the PICU: A bicenter descriptive study. Pediatr Crib Care Med 2019; 20:e400–e409
4. Jutras C, Sauthier M, Tucci M, et al.: Incidence of anemia at discharge of 4890 consecutive pediatric intensive care survivors. Vox Sanguinis 2018; 5:347
5. Bateman AP, McArdle F, Walsh TS: Time course of anemia during six months follow up following intensive care discharge and factors associated with impaired recovery of erythropoiesis. Crib Care Med 2009; 37:1906–1912
6. Warner MA, Hanson AC, Frank RD, et al.: Prevalence of and recovery from anemia following hospitalization for critical illness among adults. JAMA Netw Open 2020; 3:e2017843
7. Du Pont-Thibodeau G, Jutras C, Lacroix J: Anemia at PICU discharge: Is it a problem? Pediatr Crit Care Med 2020; 21:597–598
8. Kapoor RK, Kumar A, Chandra M, et al.: Cardiovascular responses to treadmill exercise testing in anemia. Indian Pediatr 1997; 34:607–612
9. Walter T: Effect of iron-deficiency anaemia on cognitive skills in infancy and childhood. Baillieres Clin Haematol 1994; 7:815–827
10. Yager JY, Hartfield DS: Neurologic manifestations of iron deficiency in childhood. Pediatr Neurol 2002; 27:85–92
11. Ai Y, Zhao SR, Zhou G, et al.: Hemoglobin status associated with performance IQ but not verbal IQ in Chinese preschool children. Pediatr Int 2012; 54:669–675
12. Jutras C, Charlier J, Francois T, Du Pont-Thibodeau G. Anemia in pediatric critical care. Int J Clin Transfus 2020;8:23–33
13. Sloniewsky D: Anemia and transfusion in critically ill pediatric patients: A review of etiology, management, and outcomes. Crib Care Clin 2013; 29:301–317
14. Smoller BR, Kruskall MS: Phlebotomy for diagnostic laboratory tests in adults. Pattern of use and effect on transfusion requirements. N Engl J Med 1986; 314:1233–1235
15. Smoller BR, Kruskall MS, Horowitz GL: Reducing adult phlebotomy blood loss with the use of pediatric-sized blood collection tubes. Am J Clin Pathol 1989; 91:701–703
16. Linderkamp O, Versmold HT, Riegel KP, et al.: Estimation and prediction of blood volume in infants and children. Eur J Pediatr 1977; 125:227–234
17. Orliaguet G, Lacroix J, Hébert P: Choc Hémorragique (Hemorrhagic Shock)). In: Urgences et Soins Intensifs Pédiatriques. Second edition. Lacroix J, Gauthier M, Lecrec F, et al (Eds). Paris, Elsevier-Masson, 2007, pp 167–186
18. Sauthier M, Landry-Hould F, Leteurtre S, et al.: Comparison of the Automated Pediatric Logistic Organ Dysfunction-2 versus Manual Pediatric Logistic Organ Dysfunction-2 score for critically ill children. Pediatr Crib Care Med 2020; 21:e160–e169
19. Valentine SL, Bembea MM, Muszynski JA, et al.; Pediatric Critical Care Transfusion and Anemia Expertise Initiative (TAXI); Pediatric Critical Care Blood Research Network (BloodNet), and the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network: Consensus recommendations for RBC transfusion practice in critically ill children from the pediatric critical care transfusion and anemia expertise initiative. Pediatr Crib Care Med 2018; 19:884–898
20. Lau W: Neonatal and pediatric transfusion. In: Clinical Guide to Transfusion. Chapter 13. Clarke G, Blajchman M (Eds). [online] Ottawa, Canadian Blood Services, 2017, pp 1–11
21. Peruzzi WT, Parker MA, Lichtenthal PR, et al.: A clinical evaluation of a blood conservation device in medical intensive care unit patients. Crib Care Med 1993; 21:501–506
22. Corson AH, Fan VS, White T, et al.: A multifaceted hospitalist quality improvement intervention: Decreased frequency of common labs. J Hosp Med 2015; 10:390–395
23. Kleinpell RM, Farmer JC, Pastores SM: Reducing unnecessary testing in the intensive care unit by choosing wisely. Acute Crib Care 2018; 33:1–6
24. World Health Organization: Global Forum for Blood Safety: Patient Blood Management. 2011. Available at: Accessed April 2, 2022
25. Lin JC, Strauss RG, Kulhavy JC, et al.: Phlebotomy overdraw in the neonatal intensive care nursery. Pediatrics 2000; 106:E19
26. Valentine SL, Bateman ST: Identifying factors to minimize phlebotomy-induced blood loss in the pediatric intensive care unit. Pediatr Crib Care Med 2012; 13:22–27
27. Steffen K, Doctor A, Hoerr J, et al.: Controlling phlebotomy volume diminishes PICU transfusion: Implementation processes and impact. Pediatrics 2017; 140:e20162480
28. Gleason E, Grossman S, Campbell C: Minimizing diagnostic blood loss in critically ill patients. Am J Crib Care 1992; 1:85–90
29. MacIsaac CM, Presneill JJ, Boyce CA, et al.: The influence of a blood conserving device on anaemia in intensive care patients. Anaesth Intensive Care 2003; 31:653–657
30. Mukhopadhyay A, See KC, Chan YH, et al.: Effect of a blood conservation device in patients with preserved admission haemoglobin in the intensive care unit. Anaesth Intensive Care 2011; 39:426–430
31. Rezende E, Ferez MA, Silva Junior JM, et al.: Closed system for blood sampling and transfusion in critically ill patients. Rev Bras Ter Intensiva 2010; 22:5–10
32. Thorpe S, Thomas AN: The use of a blood conservation pressure transducer system in critically ill patients. Anaesthesia 2000; 55:27–31
33. Ballin A, Livshiz V, Mimouni FB, et al.: Reducing blood transfusion requirements in preterm infants by a new device: A pilot study. Acta Paediatr 2009; 98:247–250
34. Mukhopadhyay A, Yip HS, Prabhuswamy D, et al.: The use of a blood conservation device to reduce red blood cell transfusion requirements: A before and after study. Crib Care 2010; 14:R7
35. Weber CF, Görlinger K, Meininger D, et al.: Point-of-care testing: A prospective, randomized clinical trial of efficacy in coagulopathic cardiac surgery patients. Anesthesiology 2012; 117:531–547
36. Clouzeau B, Caujolle M, San-Miguel A, et al.: The sustainable impact of an educational approach to improve the appropriateness of laboratory test orders in the ICU. PLoS One 2019; 14:e0214802
37. Jung YH, Lee J, Kim HS, et al.: The efficacy of noninvasive hemoglobin measurement by pulse CO-oximetry in neonates. Pediatr Crib Care Med 2013; 14:70–73
38. Joseph B, Haider A, Rhee P: Non-invasive hemoglobin monitoring. Int J Surg 2016; 33:254–257
39. Kazanasmaz H, Demir M: The Comparison of hemoglobin values measured by blood and continuous non-invasive monitoring (SpHb) in newborn infants. J Trop Pediatr 2021; 67:fmaa050
40. Phillips MR, Khoury AL, Bortsov AV, et al.: A noninvasive hemoglobin monitor in the pediatric intensive care unit. J Surg Res 2015; 195:257–262
41. Low LL, Harrington GR, Stoltzfus DP: The effect of arterial lines on blood-drawing practices and costs in intensive care units. Chest 1995; 108:216–219
42. Michel J, Hofbeck M, Balcells JV, et al.: Implementation of a blood-saving program in a pediatric intensive care unit to reduce diagnostic blood loss. Klin Padiatr 2020; 232:197–202
43. Delgado-Corcoran C, Bodily S, Frank DU, et al.: Reducing blood testing in pediatric patients after heart surgery: A quality improvement project. Pediatr Crib Care Med 2014; 15:756–761
44. Delgado-Corcoran C, Wolpert KH, Lucas K, et al.: Hematocrit levels, blood testing, and blood transfusion in infants after heart surgery. Pediatr Crib Care Med 2016; 17:1055–1063
45. Karanjkar A, Poonam MK, Sharan S, et al. A prospective randomized clinical trial of efficacy of algorithm-based point of care guided hemostatic therapy in cyanotic congenital heart disease surgical patients. J Cardiac Crib Care 2019;3:8–16
46. Canadian Society for Medical Laboratory Science: Choosing Wisely Canada: Medical Laboratory Science. 2020. Available at: Accessed April 2, 2022
47. Pieracci FM, Henderson P, Rodney JR, et al.: Randomized, double-blind, placebo-controlled trial of effects of enteral iron supplementation on anemia and risk of infection during surgical critical illness. Surg Infect (Larchmt) 2009; 10:9–19
48. Litton E, Baker S, Erber WN, et al.; IRONMAN Investigators; Australian and New Zealand Intensive Care Society Clinical Trials Group: Intravenous iron or placebo for anaemia in intensive care: The IRONMAN multicentre randomized blinded trial: A randomized trial of IV iron in critical illness. Intensive Care Med 2016; 42:1715–1722
49. Mesgarpour B, Heidinger BH, Roth D, et al.: Harms of off-label erythropoiesis-stimulating agents for critically ill people. Cochrane Database Syst Rev 2017; 8:CD010969
50. Ganz T: Anemia of inflammation. N Engl J Med 2019; 381:1148–1157
51. Ngo QN, Matsui DM, Singh RN, et al.: Anemia among pediatric critical care survivors: Prevalence and resolution. Crib Care Res Pract 2013; 2013:684361

anemia; blood sampling; blood testing; children; critical care; hemoglobin; iatrogenic anemia; laboratory testing; patient blood management; pediatric intensive care unit; pediatrics; phlebotomy; risk factors; transfusion

Copyright © 2022 by the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies