help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on April 17, 2008, doi:10.1164/rccm.200711-1637OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200711-1637OCv1
178/1/26    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bateman, S. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bateman, S. T.
American Journal of Respiratory and Critical Care Medicine Vol 178. pp. 26-33, (2008)
© 2008 American Thoracic Society
doi: 10.1164/rccm.200711-1637OC

Original Article

Anemia, Blood Loss, and Blood Transfusions in North American Children in the Intensive Care Unit

Scot T. Bateman1, Jacques Lacroix2, Katia Boven3, Peter Forbes4, Roger Barton5, Neal J. Thomas6, Brian Jacobs7, Barry Markovitz8, Brahm Goldstein9, James H. Hanson10, H. Agnes Li3 and Adrienne G. Randolph4 for the Pediatric Acute Lung Injury and Sepsis Investigators Network*

1 University of Massachusetts Medical Center, Worcester, Massachusetts; 2 CHU Sainte-Justine, Montreal, Canada; 3 Johnson and Johnson Pharmaceutical, Research and Development, Raritan, New Jersey; 4 Children's Hospital, Boston, Massachusetts; 5 Children's Hospital of Saint Francis, Tulsa, Oklahoma; 6 Penn State Children's Hospital, Hershey, Pennsylvania; 7 Children's National Medical Center, Washington, DC; 8 Children's Hospital of Los Angeles, Los Angeles, California; 9 Novo Nordisk, Inc., Princeton, New Jersey; and 10 Children's Hospital and Research Center Oakland, Oakland, California

Correspondence and requests for reprints should be addressed to Scot T. Bateman, M.D., University of Massachusetts Medical Center, Department of Pediatrics, H5-524, 55 Lake Avenue, North Worcester, MA 01655. E-mail: batemans{at}ummhc.org


   ABSTRACT
TOP
 ABSTRACT
AT A GLANCE COMMENTARY
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Rationale: Minimizing exposure of children to blood products is desirable.

Objectives: We aimed to understand anemia development, blood loss, and red blood cell (RBC) transfusions in the pediatric intensive care unit (PICU).

Methods: Prospective, multicenter, 6-month observational study in 30 PICUs. Data were collected on consecutive children (<18 yr old) in the PICU for 48 hours or more.

Measurements and Main Results: Anemia development, blood loss, and RBC transfusions were measured. A total of 977 children were enrolled. Most (74%) children were anemic in the PICU (33% on admission, 41% developed anemia). Blood draws accounted for 73% of daily blood loss; median loss was 5.0 ml/day. Forty-nine percent of children received transfusions; 74% of first transfusions were on Days 1–2. After adjusting for age and illness severity, compared with nontransfused children, children who underwent transfusion had significantly longer days of mechanical ventilation (2.1 d, P < 0.001) and PICU stay (1.8 d, P = 0.03), and had increased mortality (odds ratio [OR], 11.6; 95% confidence interval [CI], 1.43–90.9; P = 0.02), nosocomial infections (OR, 1.9; 95% CI, 1.2–3.0; P = 0.004), and cardiorespiratory dysfunction (OR, 2.1; 95% CI, 1.5–3.0; P < 0.001). High blood loss per kilogram body weight from blood draws (OR, 1.11; 95% CI, 1.03–1.2; P = 0.01) was associated with RBC transfusion more than 48 hours after admission. The most common indication for transfusion was low hemoglobin (42%). Pretransfusion hemoglobin values varied greatly (mean, 9.7 ± 2.7 g/dl).

Conclusions: Critically ill children are at significant risk for developing anemia and receiving blood transfusions. Transfusion in the PICU was associated with worse outcomes. It is imperative to minimize blood loss from blood draws and to set clear transfusion thresholds.

Key Words: blood loss • anemia • transfusions • pediatric • intensive care • red blood cells



   AT A GLANCE COMMENTARY
TOP
 ABSTRACT
 AT A GLANCE COMMENTARY
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Scientific Knowledge on the Subject
Recent emphasis directed toward minimizing anemia and decreasing transfusions in critically ill adult patients has led to a significant need to better understand the burden of anemia and transfusions in critically ill children.

What This Study Adds to the Field
Critically ill children are at significant risk for developing anemia and receiving blood transfusions. Transfusion in the PICU was associated with worse outcomes.

 
Anemia is common in critically ill children admitted to the pediatric intensive care unit (PICU) (1). There are numerous possible causes for the anemia of critical illness, including chronic anemia, overt and occult blood loss (2), underlying disease and treatments causing bone marrow suppression. An inadequate erythropoietin response to anemia in critically ill patients is described in adults (3–5) as well as in children (1). However, there are no data available on blood loss in children admitted to the PICU.

Red blood cell (RBC) transfusions are a common therapy in critically ill and injured children. There are multiple risks associated with RBC transfusions, including transfusion-transmitted infections, transfusion-related acute lung injury, hemodynamic compromise, intravascular volume overload, acute hemolysis, and immunosuppression (6–9). In adults, RBC transfusions have also been associated with prolongation of mechanical ventilation (10), diminished organ function, and even death (11, 12). To date, no multicenter prospective data on anemia and transfusions in the PICU are available.

In this study, we focused on children with a longer PICU stay who could most benefit from PICU interventions such as blood conservation protocols or erythropoietin therapy. Children with a PICU stay greater than 48 hours represent approximately 20% of PICU admissions, but account for disproportionately high PICU resource utilization (13). We aimed to prospectively assess the epidemiology of anemia and red cell transfusions in this population as well as to determine the causes of blood loss in the PICU. Outcomes and complications were captured to assess any association with transfusions.


   METHODS
TOP
 ABSTRACT
 AT A GLANCE COMMENTARY
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
This was a prospective, multicenter, epidemiologic, observational study conducted in 30 PICUs of the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network in the United States and Canada from September 8, 2004, to March 29, 2005. All consecutive children, younger than 18 years, admitted for any reason were eligible for study enrollment once remaining in the PICU for more than 48 hours (see the online supplement for information on children with PICU stay <= 48 h). Exclusion criteria included the following: premature neonates (corrected gestational age <= 37 wk and age < 28 d), prior participation in the survey, family history of refusing blood transfusions, involvement in other transfusion or blood management–related research, pregnancy, impending brain death, and recent (within 7 d) PICU stay of more than 72 hours. Enrollment in the study continued until the predefined sample size (~1,000 children with length of stay > 48 h) had been reached. The survey did not require additional interventions/procedures that were not part of routine medical practice. The institutional review board approved the study at each site. Written, informed consent was obtained for all enrolled subjects.

All blood loss information was collected prospectively on all children from admission onward. Other data from the first 48 hours after admission were collected retrospectively. All data after 48 hours of admission were collected prospectively. The number of blood draws and volume of each draw were recorded by the bedside nurse and collected for all children each day the patient remained in the PICU. A day was defined as a calendar day starting at midnight. Data were collected for all participants for a maximum of 28 days total or until hospital discharge, interinstitutional transfer, or death. Children readmitted to the PICU less than 48 hours after transfer were regarded as still in the PICU.

Data collected on admission included the following: demographics, severity of illness using the Pediatric Risk of Mortality (PRISM) III score (14), organ dysfunction using Pediatric Logistic Organ Dysfunction (PELOD) score (15), and multiple organ dysfunction score (MODS) (16). Daily data collection included lowest daily hemoglobin (Hb) level, RBC transfusion events, reason for physician ordering transfusion, blood loss from blood draws, PELOD and MODS variables, and clinical events including mechanical ventilation, specific technologies, surgery, or complications. Anemia was defined as an Hb concentration 2 SD below the mean Hb concentration for each age group. The cutoff values to determine anemia were as follows: 14.5 g/dl for neonates, dropping to 9 g/dl at 2 months of age, then rising to 10.5 g/dl at 6 months of age, 11.5 g/dl at 2 years of age, and 12 g/dl in females and 13 g/dl in males at adolescence (17). The following categories of severity of anemia were created for subsequent analysis to assess the impact on degree of anemia: "severe anemia," an Hb of less than 7.0 g/dl; "moderate anemia," an Hb greater than 7.0 and less than 10.0 g/dl; and "mild anemia," anemia and an Hb greater than 10 g/dl.

Chi-square tests were used to make unadjusted bivariate tests of association between the outcomes and categorical predictors. For continuous predictors, Student's t tests were used. Children with anemia on admission to the PICU, children who became anemic during their PICU stay, and children who were never anemic were compared with analysis of variance. The relationship between complications and transfusion on Day 1 or 2 was examined by comparing the complications on Day 3 or later in two groups: (1) those with one or more transfusions on Day 1 or 2 (n = 363) and (2) those with no transfusions during their PICU stay (n = 494). Logistic regression, adjusted for age and admission PRISM III score, was used to compare odds of complication for the subjects who did or did not undergo transfusion. Logistic regression was also used to model risk for developing anemia and risk for receiving a transfusion. Separate regression models were computed for each outcome. In each case, a backward variable selection procedure was used to eliminate predictors not significantly associated with the outcomes. Determination of clinical risk factors for the logistic regression models—(1) developing anemia after Day 2 and (2) getting a transfusion after Day 2—were calculated. Children receiving extracorporeal membranous oxygenation (ECMO) on Day 1 or 2 (n = 19) were excluded from regression models (SAS version 9.1; SAS Institute, Cary, NC). Median Hb for the transfused and nontransfused groups was compared by ranking Hb levels, by day, for the two transfusion groups combined and using a generalized estimating equations regression model (SAS Proc Genmod; SAS Institute) to test for a main effect of transfusion group.


   RESULTS
TOP
 ABSTRACT
 AT A GLANCE COMMENTARY
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Children were enrolled from moderate- to large-sized PICUs (70% in children's hospitals and 60% in academic centers) in the United States (25 sites, at least 1 from each U.S. Census region) and Canada (5 sites from 4 regions). Table E1 (see the online supplement) details enrollment by census region. The median number of PICU beds was 20 (range, 8–67).

There were 5,570 admissions during the study period; 1,097 (19.7%) remained in the PICU for more than 48 hours. Of these, 986 were enrolled (consent rate, 89.9%). There were nine incomplete case report forms (0.7%) from one study site, leaving 977 children for analysis. Baseline demographic characteristics of these children are presented in Table 1 together with baseline data on anemic children, children who received a transfusion, and those that developed anemia or received a late transfusion.


View this table:
[in this window]
[in a new window]

  		TABLE 1. BASELINE CHARACTERISTICS OF THE CHILDREN ON ADMISSION TO THE PEDIATRIC INTENSIVE CARE UNIT

 
In the youngest age category (<=28 d), 63% of patients were primarily admitted as surgical nontrauma, whereas in the three oldest age groups, the highest proportion of children were admitted as medical nontrauma. Children in the age category ranging from 28 days to less than 2 years were equally distributed between these two admission types. The most common primary PICU admitting diagnostic category for children younger than 28 days was the cardiovascular system (n = 62 [75%]). In contrast, the respiratory system was most common among the three middle age categories (38–42% of children). In the oldest age group, the respiratory and central nervous systems were the most common admitting diagnostic categories (32 and 34% of children, respectively). There was an increasing trend with age in the proportion of children whose primary PICU admitting diagnostic category was the central nervous system.

Anemia
A history of anemia within 7 days before PICU admission was reported in 15% of children. Anemia in the PICU was common: it was present on admission in 33% of children. An additional 41% became anemic during their PICU stay. Only 26% of the children never became anemic. Anemia overall (either on admission or during PICU stay) was more common in neonates and adolescents (Figure 1) (P < 0.0001).


Figure 1
View larger version (45K):
[in this window]
[in a new window]

  		Figure 1. Proportion of patients with anemia overall (patients with anemia on admission coupled with those that developed anemia in the pediatric intensive care unit [PICU]) and of patients who received at least one red blood cell transfusion during their PICU stay. *Groups were statistically different by age (P < 0.0001).

 
Children with anemia on admission to the PICU had a higher severity of illness score with higher PRISM III (mean, 5.3 ± 5.8) than children who became anemic during their PICU stay (mean, 3.8 ± 5.1) and children who were never anemic (mean, 3.3 ± 4.5) (P < 0.0001). There was no difference in the amount of average daily blood loss from all sources among these three groups. The PICU length of stay was greatest in the group who developed anemia in the PICU (mean, 10.4 ± 7.8 d) compared with those anemic on admission (8.9 ± 7.0 d) or never anemic (6.6 ± 5.9 d) (P < 0.0001). In addition, children who developed anemia in the PICU had more days of mechanical ventilation (mean, 6.7 d) than those anemic on admission (5.7 d) or never anemic (2.7 d) (P < 0.0001).

A multivariate logistic regression model was used to identify independent risk factors fore children developing anemia more than 48 hours after admission. Children were excluded if they were receiving ECMO or if they had anemia on admission or anemia on Days 1–2 (438 children were left for analysis; 176 developed anemia). Factors predictive of anemia development in the univariate analyses were tested in the multivariate model and are reported in Table 2. Significant predictors of initial anemia development more than 48 hours after PICU admission were age of 28 days or younger, no RBC transfusion on PICU Days 1–2, presence of shock on PICU admission, admission category of "other" (gastrointestinal, endocrine, renal, hematology/oncology), baseline PELOD score of 11 or more, and having a respiratory comorbid condition. Factors not predictive of developing later anemia in the bivariate analysis were transfusion before admission, sex, race, admission type (medical, surgical nontrauma, trauma), PRISM III score (0, 1–5, >=6), and blood loss/kg from blood draws on admission for Days 1–2.


View this table:
[in this window]
[in a new window]

  		TABLE 2. PREDICTORS OF DEVELOPMENT OF ANEMIA AFTER PEDIATRIC INTENSIVE CARE UNIT DAY 2 (n = 438)*

 
Blood Loss
Almost all children (96.5%) had blood loss from blood draws, 325 (33%) had some blood loss due to procedures (median daily loss was 0.25 ml/kg), and 233 (24%) had blood loss due to spontaneous bleeding (median daily loss was 2.56 ml/kg). Blood loss from blood draws accounted for the majority of total blood loss during the ICU stay in all age groups (mean, 72.9%; median, 100%). Children who had either an arterial line or a central venous catheter in place during the PICU stay had a 2.3- to 4-times higher median number of blood draws per day through Day 14 of their PICU stay than children with peripheral venous lines only.

The mean volume of blood loss per blood draw was 2.7 ± 2.3 ml/draw (median, 2.0 ml/draw). There were a mean of three blood draws/day with a mean daily volume of blood loss of 8.25 ± 21.5 ml/day (median, 5.0 ml/d; adjusted for weight, 0.32 ml/kg/d). An analysis of daily blood loss from blood draws by age category adjusted for weight (kg) shows an inverse relationship between blood loss/kg and age (P = 0.02) (Figure 2). For all children, there was a decreasing trend in the median number of daily blood draws per patient over time during the PICU stay, with the highest number of blood draws per patient (mean, 7; median, 6) occurring on the second calendar day of admission.


Figure 2
View larger version (28K):
[in this window]
[in a new window]

  		Figure 2. Average number of blood draws per day per patient and average volume of blood collected per day per patient by age of patients. *Groups were statistically different by age (P < 0.02).

 
RBC Transfusion
Overall, 475 (49%) children received one or more RBC transfusions during the PICU stay and 6% received a transfusion after PICU discharge. The first transfusion was given within 48 hours of PICU admission 74% of the time (22% received their first transfusion between Days 2 and 7, and only 4% received their first transfusion > 7 d after admission to PICU). One hundred eighty-one (38%) children who received a transfusion had only one transfusion, whereas 110 (23%) had two, 50 (11%) had three, and 34 (7%) had four transfusions. The remaining 100 (21%) had four or more transfusions, and this group received 1,017 (50%) of all transfusion events. The time course of transfusions is presented in Figure 3, which shows the predominance of early transfusions. Transfusion-related complications of fever, hemolysis, or transfusion reaction were rare at 0.5%.


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Figure 3. Cumulative distribution of day of transfusion for children with one, two, and three total transfusions.

 
Data on packed RBC (PRBC) characteristics were available for 1,301 transfusions that occurred in the PICU after enrollment. This revealed that 69% of these PICU transfusions were with irradiated blood, 92% were from unrelated donors, and 86% were leukocyte depleted. The storage age of the blood was recorded in 1,288 transfusions. Sixty-five percent (842) of the PRBCs given were less than or equal to 14 days of storage. The mean age of the blood, when recorded, was 13.3 days.

Compared with the subset of children who did not receive a transfusion during the PICU stay, the children who received a transfusion in the PICU were younger (mean age, 4.5 vs. 6.6 yr; P < 0.001), had a higher rate of anemia on admission (44 vs. 22%, P < 0.001), and had higher baseline PRISM III score (P < 0.001). Transfusion incidence by age is presented in Figure 1. Children who received at least one RBC transfusion during their PICU stay also had a greater average daily blood loss (median, 1.6 vs. 0.2 ml/kg; P < 0.001), a higher incidence of anemia (24 vs. 7%, P < 0.001) and transfusions (41 vs. 2%, P < 0.001) within 7 days before PICU admission, more surgical/invasive procedures at PICU admission (47 vs. 38%, P = 0.003), and more shock on admission (21 vs. 6%, P < 0.001). Most of the children who had a primary admitting diagnosis pertaining to the cardiovascular system received a transfusion during the PICU stay (74%).

A total of 3,521 predefined complications were documented during the PICU stay. The effect of transfusion on outcomes was evaluated by comparing PICU complications that occurred after 48 hours in those children who had received a transfusion on Day 1 or 2 (n = 363) with those who never received a transfusion (n = 494) during their PICU stay. Correcting for age and for admitting PRISM III scores, the transfused group had an increased risk of death (odds ratio [OR], 11.6; 95% confidence interval [CI], 1.43–90.9; P = 0.02), an increased risk of death and/or cardiac arrest (OR, 20.0; 95% CI, 2.6–166.7; P = 0.004), a higher rate of nosocomial infections (OR, 1.9; 95% CI, 1.2–3.0; P = 0.004), and more cardiac or respiratory dysfunction (OR, 2.1; 95% CI, 1.5–3.0; P < 0.001). The transfused group had a longer PICU length of stay (7.5 d in nontransfused vs. 9.3 d in transfused, P = 0.0002) and, for children ventilated beyond 48 hours (n = 421 [204 nontransfused and 217 transfused]), a longer length of mechanical ventilation (7.5 d in nontransfused vs. 9.6 d in transfused, P = 0.003). Of note, 12 of 15 deaths (80%) in the transfused group occurred in patients who had received more than four transfusions in the PICU.

The number of patients who received a transfusion after 48 hours was 162 (17%). Predictors for receiving an RBC transfusion after the first 2 days in the PICU were calculated using a multivariate logistic regression analysis, excluding only those children receiving ECMO on PICU Days 1–2 (n = 19). Factors predictive of a later transfusion in the bivariate analysis were tested in the multivariate model and are reported in Table 3. Significant predictors of receiving a transfusion more than 48 hours after PICU admission were age of 28 days or younger, RBC transfusion on PICU Days 1 or 2, presence of severe or moderate anemia, presence of shock on PICU admission, admission because of trauma, admission category of cardiovascular or "other," baseline PRISM III score of more than 5, having a hematologic/oncologic comorbid condition, and high mean daily volume/kg of blood loss from blood draws. Factors not predictive in the univariate analysis were sex, race, and admission type (medical, surgical nontrauma, trauma).


View this table:
[in this window]
[in a new window]

  		TABLE 3. PREDICTORS OF TRANSFUSION AFTER PEDIATRIC INTENSIVE CARE UNIT DAY 2 (n = 958)*

 
The reason for transfusion in the PICU was reported by the prescribing physician and is presented in Table 4. Low Hb was the most common reason listed (42%), and this category had a significantly lower mean pretransfusion Hb than the other groups (P < 0.001). Bone marrow suppression and specific technologies were infrequent primary reasons for transfusion, but those patients had many more transfusion events per patient. Mean number of PICU transfusions varied significantly between reason groups (P < 0.0001).


View this table:
[in this window]
[in a new window]

  		TABLE 4. REASON FOR FIRST TRANSFUSION BY ORDERING PHYSICIAN (n = 476)

 
There was marked variability in Hb values before transfusion. Table 5 shows the mean and median pretransfusion Hb plus intraquartile ranges for the first transfusion by age, admitting type (medical vs. surgical/trauma), and severity of illness (PRISM III score). Younger children had a higher mean pretransfusion Hgb (P < 0.0001). Children with a lower PRISM III score (<=5) and a medical admission type had lower pretransfusion Hb levels (P < 0.001).


View this table:
[in this window]
[in a new window]

  		TABLE 5. PRETRANSFUSION HEMOGLOBIN VALUES

 
Median daily Hb values were lower for children who received a transfusion compared with those who did not (P < 0.0001; Figure 4). Excluding neonates, Hb values for 75% of children for the first 14 days of ICU stay were 8.4 g/dl or greater in those who did not receive a transfusion and 8.1 g/dl or more in those who received one or more PRBC transfusion.


Figure 4
View larger version (10K):
[in this window]
[in a new window]

  		Figure 4. Median hemoglobin by day and transfusion group (age > 2 and < 18 yr). Mean daily hemoglobin for the first 14 days was significantly less in patients who received a transfusion (P < 0.001).

 

   DISCUSSION
TOP
 ABSTRACT
 AT A GLANCE COMMENTARY
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
This is the first large, multicenter, prospective study of anemia, blood loss, and transfusion practices in critically ill children. Anemia was a common problem, affecting 74% of these children during or immediately before PICU admission. Blood loss via blood draws was particularly significant in the younger age groups. Almost half of the study population received at least one RBC transfusion. The majority (74%) received their first transfusion in the first 2 days after admission and only 4% received it after the first week. These data provide evidence against prophylactic therapy with erythropoietin to prevent blood transfusions in the PICU setting, and focus more emphasis on blood loss prevention. After correction for age and illness severity, the increased length of stay and complications shown in the children who received a transfusion provide supportive evidence that efforts at minimizing transfusions should continue.

Most data available on incidence of anemia and transfusion practices in the children come from neonates, which do not provide clear guidance for the highly variable PICU population (18–20). Anemia in critically ill children is almost always treated by blood transfusion, and transfusions have been associated with increased PICU utilization (21). Published reports suggest considerable variability in practice. A single-center Canadian study revealed that 15% of all children in the ICU received at least one transfusion (22), whereas a British single-center study reported that 48% of all children in the PICU received at least one transfusion (23). Surveys of physician opinion suggest wide variability of Hb values used to justify RBC transfusion decisions in children (24–26). Our multicenter international study showed that the proportion of children who received at least one RBC transfusion while they were in the PICU was 49%.

The rate of anemia observed in our study population was higher than expected. These data mimic the results reported on adults by Vincent and colleagues in their prospective blood loss and transfusion survey in European ICUs, with similar rates of anemia on admission (33 vs. 36%) (11). However, the prevalence of anemia in the PICU is greater due to the additional 41% who developed anemia in the PICU.

The impact of anemia on the outcome of critically ill children is not well understood. Some data suggest that severe anemia may be detrimental to critically ill children with septic shock or hemodynamic compromise (27–30). Anemia was associated with a worse outcome in the Corwin study, a descriptive prospective study that included 4,892 consecutive critically ill adults from 213 American ICUs (31). The same association was observed in 3,534 adults from 146 Europeans ICUs (11). The postoperative risk of death increases significantly when Hb concentration drops below 4 g/dl (32). Three prospective studies run in Kenya involving, respectively, 2,433, 1,223, and 1,269 hospitalized children showed that the risk of death was significantly higher if their Hb concentration was lower than 5 g/dl and if they did not receive an RBC transfusion; these children were not critically ill, but most had respiratory symptoms (33–35). Therefore, severe anemia seems to be an independent risk factor for death in sick patients, at least when the Hb concentration drops below 5 g/dl. On the other hand, a large multicenter randomized clinical trial, the TRIPICU (Transfusion Requirements in the Pediatric Intensive Care Unit) study, showed that a restrictive RBC transfusion strategy in which the threshold Hb concentration to prescribe a transfusion was 7 g/dl was not inferior to a liberal strategy with a threshold of 9.5 g/dl in stable, critically ill children (36). This suggests that an RBC transfusion is probably not useful in stable children if their Hb concentration is above 7 g/dl. The Hb concentration that should prompt pediatric intensivists to prescribe an RBC transfusion in unstable children remains to be determined. Higher Hb concentrations may be required in children with greater severity of illness (e.g., shock) and in specific subpopulations (e.g., postoperative cardiac surgery).

Small body size and small total blood volume make children vulnerable to anemia secondary to blood draws despite the use of microsampling techniques. Our findings point out the need for future studies of blood conservation strategies via phlebotomy in the PICU. The median blood volume loss was 5.0 ml/day in our study, which is markedly less than the 41 ml/day reported by Vincent and colleagues in adult ICUs (11). However, the degree of blood volume loss remains significant, particularly because almost half of our population was younger than 24 months. The total blood volume of a 5-kg child is about 400 ml in comparison to 5 L in a typical adult. Younger children have higher circulating blood volumes per kilogram (80 ml/kg) than older children and adults (70 ml/kg) (37), but the burden on them from iatrogenic blood loss remains significantly higher due to their overall lower blood volumes. Blood loss was not predictive of later anemia, most likely because such a high percentage of patients were anemic (74%) and our model did not determine whether blood draws led to more severe anemia. More importantly, we noted that blood loss secondary to blood draws early in admission was predictive of later transfusion.

Vincent and coworkers reported a transfusion rate of 34% in their adult trial. Our findings of a 49% transfusion rate are higher, potentially due to a different transfusion threshold practiced in the PICU. The mean pretransfusion Hb in the adult trial was 8.4 ± 1.3 g/dl compared with our findings of 9.7 ± 2.7 g/dl (11). The results of the TRIPICU study suggest that it is safe not to give RBC transfusion to stable, critically ill children if their Hb concentration is higher than 7 g/dl (35). In our study, the children who could be considered more stable (based on low PRISM III scores < 5) and who received a transfusion (n = 253) had a mean pretransfusion Hb of 9.23 ± 2.4 g/dl. Markedly fewer children in these groups might receive a transfusion if the results of the TRIPICU study are applied.

Almost three-quarters of our patients received their transfusions in the first 2 days. Unless the criteria for RBC transfusion changes, it is unlikely that therapies such as erythropoietin will be of benefit to prevent transfusions given the time required for this therapy to affect a change in Hb in critically ill patients (38). The use of erythropoietin in the PICU may merit further study, however, because of recent evidence demonstrating a mortality benefit in critically ill adults, which was independent of its effect on Hb (39).

This study represents the largest assessment of the RBC transfusion rationale among physicians caring for critically ill patients. Low Hb was almost twice as common as blood loss as the primary reason for transfusion. Therefore, clearer guidelines for Hb values that should trigger a transfusion would benefit clinicians. On the other hand, our data suggest that some attention must also be paid to other justifications, such as the severity of illness and blood loss.

The strengths of our study include the size of the study population, and the focus on the relationship between PICU day and transfusion risk. We noted that those children staying in the PICU more than 48 hours are at higher risk for transfusion. The proportion of children who received an RBC transfusion was 17% in 216 children who stayed in the PICU less than 2 days in comparison to 49% in the 977 who stayed more than 48 hours (see Table E2). The generalizability to the entire PICU population must account for the fact that only approximately 20% of PICU patients have a length of stay greater than 48 hours.

In this long-stay population, our data show that, when corrected for severity of illness, transfusions were associated with significantly worse outcomes. This has not been shown previously in any prospective study in the PICU and warrants a serious look into the use of transfusions in this population. Our data are consistent with the adult literature by Vincent and colleagues and Corwin and coworkers, which also showed longer length of stay and increased morbidity and mortality in the transfused ICU patients (11, 31). However, this was contrasted by the more recent analysis of adult ICU patients, which showed no mortality increase in patients who received a transfusion (40). This discrepancy has been suggested to be due to PRBC factors such as leukoreduction and older age of blood. The storage age of blood of greater than 2 weeks was reported to be an independent risk factor of mortality in adult cardiac surgical patients (41). Our findings of worse outcomes in the PICU children who received a transfusion, however, were with blood that had an 86% leukoreduction rate and 65% with storage age less than 14 days. Therefore, the effect of transfusion in our population appears to be independent of the suspected PRBC factors. A single-center retrospective study of transfusions in critically ill children by Kneyber and colleagues also found increased mortality and morbidity in children who received a transfusion, and all of their transfusions were with leukoreduced blood (42). The finding that a majority of the deaths in this study occurred in patients with more than four transfusions raises the need to better understand the effects of transfusion in unstable children. There was no increased death rate in the children who received a single or double transfusions. We did confirm the findings of Slonim and coworkers that complications directly related to the administration of PRBCs in children are rare (43).

In conclusion, the burden of anemia, blood loss, and transfusions in the PICU population are significant. Efforts to develop guidelines are clearly needed. Prospective studies taking into account the data provided in this large multicenter epidemiologic study should be undertaken to estimate the clinical impact of measures aiming to decrease blood draws, to prevent or treat anemia, and to decrease transfusions for critically ill children.


   FOOTNOTES
 
Supported by Johnson and Johnson Pharmaceutical Research and Development, L.L.C.

* A list of participating PALISI investigators appears at the end of this article. Back

This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org

Originally Published in Press as DOI: 10.1164/rccm.200711-1637OC on April 17, 2008

Conflict of Interest Statement: S.T.B. served as a consultant for Johnson and Johnson from 2004 to 2006, total of $25,000 for 3 years. J.L. received $25,000 as a consultant for the research project financed by Johnson and Johnson. K.B. is a full-time employee of Johnson and Johnson. P.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. N.J.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. B.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. J.H.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. H.A.L. is an employee of Johnson and Johnson. A.G.R. served as a consultant for Johnson and Johnson during the study.

Participating PALISI investigators in the United States and Canada are as follows: Shamile Abd-Allah, M.D., Loma Linda University Children's Hospital, Loma Linda, CA; Roger Barton, M.D., Children's Hospital of Saint Francis, Tulsa, OK; Scot Bateman, M.D., Adrienne Randolph, M.D., Heather Healey, R.N., Children's Hospital, Boston, MA; James Besunder, M.D., Akron Children's Hospital, Akron, OH; Jeffrey Blumer, M.D., Rainbow Babies and Tod Hospital, Cleveland, OH; Ira Cheiftez, M.D., Andora Bass, M.D., Duke Medical Center, Durham, NC; Heidi Dalton, M.D., Children's National Medical Center, Washington, DC; Emily Dobyns, M.D., Children's Hospital, Denver, CO; Jill Fitch, M.D., Children's Hospital, Columbus, OH; Rainer Gedeit, M.D., Children's Hospital Wisconsin, Milwaukee, WI; Brahm Goldstein, M.D., Doernbecher Children's Hospital, Oregon Health and Science University, Portland, OR; Marek Grzeszczak, M.D., Vanderbilt Children's Hospital, Nashville, TN; James H. Hanson, M.D., Children's Hospital and Research Center Oakland, Oakland, CA; Rashed Hasan, M.D., Hurley Medical Center, Flint, MI; Brian Jacobs, M.D., Cincinnati Children's Hospital, Cinncinati, OH; Larry Jefferson, M.D., Baylor College of Medicine, Houston, TX; Daniel Levin, M.D., Dean Jarvis, R.N., Children's Hospital at Dartmouth, Hanover, NH; Barry Markovitz, M.D., Washington University, St. Louis, MO; Christopher Newth, M.D., Lori Auw, R.C.T., Children's Hospital of Los Angeles, Los Angeles, CA; Vipul Patel, M.D., Dayton Children's Hospital, Dayton, OH; Steven Pon, M.D., Charlene S. Correa, B.A., New York Presbyterian Hospital, New York, NY; Edward Seferian, M.D., Mayo Clinic, Rochester, MN; Neal J. Thomas, M.D., Joseph Hess, R.N., Jennifer Stokes, R.N., Penn State Children's Hospital, Hershey, PA; Douglas Willson, M.D., University of Virginia, Charlottesville, VA; Peter Cox, M.D., Hospital for Sick Children, Toronto, Canada; Ari Joffe, M.D., University of Alberta, Edmonton, Canada; Jacques Lacroix, M.D., Hospital Sainte-Justine, Montreal, Canada; Kusum Menon, M.D., Children's Hospital of Eastern Ontario, Ottowa, Canada; David Wensley, M.D., Children's and Women's Hospital, Vancouver, Canada.

Received in original form November 6, 2007; accepted in final form April 11, 2008


   REFERENCES
TOP
 ABSTRACT
 AT A GLANCE COMMENTARY
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Krafte-Jacobs B, Levetown ML, Bray GL, Ruttiman UE, Pollack MM. Erythropoietin response to critical illness. Crit Care Med 1994;22:821–826.[Medline]
  2. Burnum JF. Medical vampires. N Engl J Med 1983;314:1250–1251.
  3. Von Ahsen N, Muller C, Serke S, Frei U, Eckhardt KU. Important role of nondiagnostic blood loss and blunted erythropoietic response in the anemia of medical intensive care patients. Crit Care Med 1999;27:2630–2639.[CrossRef][Medline]
  4. Rogiers P, Zhang H, Leeman M, Nagler J, Neels H, Mélot C, Vincent JL. Erythropoietin response is blunted in critically ill patients. Intensive Care Med 1997;23:159–162.[CrossRef][Medline]
  5. Corwin HL, Rodriguez RM, Pearl RG, Corwin MJ, Enny C, Gettinge A. Erythropoietin response in critically ill patients [abstract]. Crit Care Med 1997;25(Suppl):A82.
  6. Blajchman MA, Vamvakas EC. The continuing risk of transfusion-trasmitted infections. N Engl J Med 2006;13:1303–1305.
  7. Blumberg N, Heal JM. Immunomodulation by blood transfusion: an evolving scientific and clinical challenge. Am J Med 1996;101:299–308.[CrossRef][Medline]
  8. Bordin JO, Heddle NM, Blajchman MA. Biologic effects of leucocytes present in transfused cellular blood products. Blood 1994;84:1703–1721.[Abstract/Free Full Text]
  9. Van der Watering LMG, Hermans J, Houbiers JGA, van den Broek PJ, Bouter H, Boer F, Harvey MS, Huysmans HA, Brand A. Beneficial effect of leukocyte depletion of transfused blood on postoperative complications in patients undergoing cardiac surgery: a randomized clinical trial. Circulation 1998;97:562–568.[Abstract/Free Full Text]
  10. Hébert PC, Blajchman MA, Cook DJ, Yetisir E, Wells G, Marshall J, Schweitzer I; Transfusion Requirements in Critical Care Investigators for the Canadian Critical Care Trials Group. Do blood transfusions improve outcomes related to mechanical ventilation? Chest 2001;119:1850–1857.[CrossRef][Medline]
  11. Vincent JL, Baron J, Reinhart K, Gattinoni L, Thijs L, Webb A, Meier-Hellmann A, Nollet G, Peres-Bota D; ABC (Anemia and Blood Transfusion in Critical Care) Investigators. Anemia and blood transfusion in critically ill patients. JAMA 2002;288:1499–1507.[Abstract/Free Full Text]
  12. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot MM, Duh MS, Shapiro MJ. Anemia and blood transfusion in the critically ill: current clinical practice in the US—the CRIT study [abstract]. Crit Care Med 2002;29(Suppl)12;A2.
  13. Marcin JP, Slonim AD, Pollack MM, Ruttimann UE. Long-stay patients in the pediatric intensive care unit. Crit Care Med 2001;29:652–657.[CrossRef][Medline]
  14. Pollack MM, Kantilala MP, Ruttiman UE. PRISM III: An updated pediatric risk of mortality score. Crit Care Med 1996;24:743–752.[CrossRef][Medline]
  15. Leteurtre S, Martinot A, Duhamel A, Proulx F, Grandbastien B, Cotting J, Gottesman R, Joffe A, Pfenninger J, Hubert P, et al. Validation of the Paediatric Logistic Organ Dysfunction (PELOD) score: prospective, observational, multicentre study. Lancet 2003;362:192–197.[CrossRef][Medline]
  16. Marshall JC, Cook DJ, Christou NV, Bernard GR, Sprung CL, Sibbald WJ. Multiple organ dysfunction score: a reliable descriptor of a complex clinical outcome. Crit Care Med 1995;23:1638–1652.[CrossRef][Medline]
  17. Miller DR. Normal blood values from birth through adolescence. In: Miller DR, Behner RL, Miller LP, editors. Blood diseases of infancy and childhood, 7th ed. St. Louis, MO: Mosby; 1995. pp. 30–53.
  18. Sacher RA, Strauss RG, Luban NL, Feil M, Anstall HB, Barnes A Jr, Blanchette VS, Butch SH, Hume HA, Kevy SV, et al. Blood component therapy during the neonatal period: a national survey of red cell transfusion practice, 1985. Transfusion 1990;30:271–276.[CrossRef][Medline]
  19. Levy GJ, Strauss RG, Hume H, Schloz L, Albanese MA, Blazina J, Werner A, Sotelo-Avila C, Barrasso C, Blanchette V, et al. National survey of neonatal transfusion practices: I. Red blood cell therapy. Pediatrics 1993;91:523–529.[Abstract/Free Full Text]
  20. Hume H, Blanchette V, Strauss RG, Levy GJ. A survey of Canadian neonatal blood transfusion practices. Transfus Sci 1997;18:71–80.[CrossRef][Medline]
  21. Goodman AM, Pollack MM, Patel KM, Luban NL. Pediatric red blood cell transfusions increase PICU resource use. J Pediatr 2003;142:95–97.[CrossRef][Medline]
  22. Armano R, Gauvin F, Ducruet T, Hume H, Lacroix J. Determinants of red blood cell transfusions in a pediatric critical care unit: a prospective descriptive epidemiological study. Crit Care Med 2005;33:2637–2644.[CrossRef][Medline]
  23. Morris KP, Naqvi N, Davies P, Smith M, Lee PW. A new formula for blood transfusion volume in the critically ill. Arch Dis Child 2005;90:724–728.[Abstract/Free Full Text]
  24. Zimmerman R, Handtrack D, Zingsem J, Weisbach V, Neidhardt B, Glaser A, Eckstein R. A survey of blood utilization in children and adolescents in a German university hospital. Transfus Med 1998;8:185–194.[CrossRef][Medline]
  25. Laverdière C, Gauvin F, Hébert PC, Rivard CI, Hume H, Toledano BJ, Lacroix J; for the Canadian Critical Care Trials Group. Survey of transfusion practices in pediatric intensive care units. Pediatr Crit Care Med 2002;3:335–340.[CrossRef][Medline]
  26. Desmet L, Lacroix J. Transfusion in pediatrics. Crit Care Clin 2004;20:299–311.[CrossRef][Medline]
  27. Orliaguet G, Gauvin F, Hume H, Lacroix J, Hébert P. Choc hémorragique [in French]. In: Lacroix J, Gauthier M, Hubert P, Leclerc F, Gaudreault P, editors. Urgences et soins intensifs pédiatriques, 2nd ed. Montréal and Paris: Les Éditions de l'Hôpital Sainte-Justine & Masson; 2007. pp. 167–186.
  28. Cropp GJA. Cardiovascular function in children with severe anemia. Circulation 1969;39:775–784.[Abstract/Free Full Text]
  29. Lucking SE, Williams TM, Chaten FC, Metz RI, Mickell JJ. Dependence of oxygen consumption on oxygen delivery in children with hyperdynamic septic shock and low oxygen extraction. Crit Care Med 1990;18:1316–1319.[Medline]
  30. Mink RB, Pollack MM. Effect of blood transfusion on oxygen consumption in pediatric septic shock. Crit Care Med 1990;18:1087–1091.[Medline]
  31. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Abraham E, MacIntyre NR, Shabot MM, Duh MS, Shapiro MJ. The CRIT study: anemia and blood transfusion in the critically ill–current clinical practice in the United States. Crit Care Med 2004;32:39–52.[CrossRef][Medline]
  32. Carson JL, Noveck H, Berlin JA, Gould SA. Mortality and morbidity in patients with very low postoperative Hb levels who decline blood transfusion. Transfusion 2002;42:812–818.[CrossRef][Medline]
  33. Lackritz EM, Campbell CC, Ruebush TK 2nd, Hightower AW, Wakube W, Steketee RW, Were JB. Effect of blood transfusion on survival among children in a Kenyan hospital. Lancet 1992;340:524–528.[CrossRef][Medline]
  34. Lackritz EM, Hightower AW, Zucker JR, Ruebush TK 2nd, Onudi CO, Steketee RW, Were JB, Patrick E, Campbell CC. Longitudinal evaluation of severely anemic children in Kenya: the effect of transfusion on mortality and hematologic recovery. AIDS 1997;11:1487–1494.[Medline]
  35. English M, Ahmed M, Ngando C, Berkley J, Ross A. Blood transfusion for severe anaemia in children in a Kenyan hospital. Lancet 2002;359:494–495.[CrossRef][Medline]
  36. Lacroix J, Hébert PC, Hutchison JH, Hume H, Tucci M, Ducruet T, Gauvin F, Collet JP, Toledano BJ, Robillard P, et al.; for the TRIPICU Investigators, the Canadian Critical Care Trials Group, and the Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network. Transfusion strategies for patients in pediatric intensive care units. N Engl J Med 2007;356:1609–1619.[Abstract/Free Full Text]
  37. Linderkamp O, Versmold HT, Riegel KP, Betke K. Estimation and prediction of blood volume in infants and children. Eur J Pediatr 1977;125:227–234.[CrossRef][Medline]
  38. Corwin HL, Gettinger A, Pearl RG, Fink MP, Levy MM, Shapiro MJ, Corwin MJ, Colton T; EPO Critical Care Trials Group. Efficacy of recombinant human erythropoietin in critically ill patients: a randomized controlled trial. JAMA 2002;288:2827–2835.[Abstract/Free Full Text]
  39. Corwin HL, Gettinger A, Fabian TC, Addison M, Pearl RG, Heard S, An R, Bowers PJ, Purton P, Klausner MA, et al. Efficacy and safety of epoetin alfa in critically ill patients. N Engl J Med 2007;357:965–976.[Abstract/Free Full Text]
  40. Vincent JL, Sakr Y, Sprung C, Harboe S, Damas P; Sepsis Occurrence in Acutely Ill Patients (SOAP) Investigators. Are blood transfusions associated with increased mortality rates? Results of the Sepsis Occurrence in Acutely Ill Patients (SOAP) study. Anesthesiology 2008;108:31–39.[Medline]
  41. Koch CG, Li L, Sessler DI, Figueroa P, Hoeltge GA, Mihaljevic T, Blackstone EH. Duration of red-cell storage and complications after cardiac surgery. N Engl J Med 2008;358:1229–1239.[Abstract/Free Full Text]
  42. Kneyber MC, Hersi MI, Twisk JW, Markhorst DG, Plotz FB. Red blood cell transfusion in critcally ill children is independently associated with increased mortality. Intensive Care Med 2007;33:1414–1422.[CrossRef][Medline]
  43. Slonim AD, Joseph JG, Tureen WM, Sharangpani A, Luban NLC. Blood transfusions in children: a multi-institutional analysis of practices and complications. Transfusion 2008;48:73–80.



This article has been cited by other articles:


Home page
 Proc Am Thorac SocHome page
K. R. Flaherty
Clinical Year in Review IV: Pediatric Pulmonary and Critical Care, Cystic Fibrosis, Asthma, and Pleural Disease
Proceedings of the ATS, September 15, 2009; 6(6): 506 - 511.
[Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
R. A. Fowler, N. K. J. Adhikari, D. C. Scales, W. L. Lee, and G. D. Rubenfeld
Update in Critical Care 2008
Am. J. Respir. Crit. Care Med., May 1, 2009; 179(9): 743 - 758.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200711-1637OCv1
178/1/26    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bateman, S. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bateman, S. T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2008 American Thoracic Society