Supplementary Online Content

Transcription

1 Supplementary Online Content Dhabangi A, Ainomugisha B, Cserti-Gazdewich C, et al. Effect of transfusion of red blood cells with longer vs shorter storage duration on elevated blood lactate levels in children with severe anemia: the TOTAL randomized clinical trial. JAMA. doi: /jama eappendix 1. Estimated RBC Dose as a Proportion of Recipient Red Cell Mass eappendix 2. Supplemental Oxygen Administration and Cerebral Tissue Oxygen Saturation eappendix 3. Resolution of Impaired Consciousness and Respiratory Distress eappendix 4. Vital Signs, Arterial Oxygen Saturation, and Electrolytes eappendix 5. Adverse Events eappendix 6. Subgroup Analysis: Patients Receiving 2 Doses of RBCs eappendix 7. Posthoc Analyses efigure 1. Transfused RBCs as a Percentage of Recipient Pretransfusion Red Cell Mass Among Those Transfused With 1 Dose (10 ml/kg) or 2 Doses (20 ml/kg) of RBCs efigure 2. An Example of a Cerebral Tissue Oxygen Saturation Curve efigure 3A. Percentage of Participants in Each Group With Stupor or Coma Observed at Different Time Points During the Study efigure 3B. Percentage of Participants in Each Group With Respiratory Distress Observed at Different Time Points During the Study efigure 4A. Vital Signs and Arterial Oxygen Saturation Among Participants in Each Group Measured at Different Time Points During the Study efigure 4B. Electrolytes and Creatinine Among Participants in Each Group Measured at Different Time Points During the Study efigure 5A. Mean Blood Lactate Values in the 2 Study Groups Among the Sub- Group of Participants Transfused With 20 ml/kg of RBCs efigure 5B. Median Cerebral Tissue Oxygen Saturation at the Start and the End of Transfusion for the 2 Study Groups Among the Subgroup of Participants Transfused With 20 ml/kg of RBCs etable 1. Number of Participants Receiving Supplemental Oxygen at Different Time Points During the Study etable 2. Vital Signs, Electrolytes, BUN, and Creatinine in the Subgroup of Participants Transfused With 20 ml/kg of RBCs etable 3. Patient Characteristics Among Those Achieving or Not Achieving a Lactate of 3 mm or Less at Hour 8 of the Study

2 This supplementary material has been provided by the authors to give readers additional information about their work.

3 eappendix 1. Estimated RBC dose as a proportion of recipient red cell mass Given the severity of anemia, each transfusion represented a substantial fraction of the recipients total red cell mass. Using the measured hemoglobin concentration of each unit and the measured volume of blood infused, we calculated the grams of hemoglobin infused. Using the recipient s measured weight and pretransfusion hemoglobin concentration, we estimated the recipient circulating red cell mass in grams of hemoglobin. The graph shows the transfused red cell content (grams of hemoglobin) as a proportion of the recipient red cell mass (grams of hemoglobin). See efigure 1.

4 efigure 1. Transfused RBCs as a percentage of recipient pretransfusion red cell mass for those receiving 1 dose (10 ml/kg) or 2 doses (20 ml/kg) of RBCs Bars show mean and standard deviation.

5 eappendix 2. Supplemental Oxygen Administration and Cerebral tissue oxygen saturation Finger oximetry measurements and use of supplemental oxygen was recorded in every patient at hour 0, 2, 4, 6, 8, and 24. Per written protocol, supplemental oxygen was given to patients with an arterial oxygen saturation <95% as measured by finger oximetry. Supplemental oxygen was given either by nasal prongs or by face mask. No patient received mechanical ventilation. Use of supplemental oxygen is shown in etable 1. Values in the table refer to the number of patients receiving the treatment listed in each row..

6 etable 1. Number of participants receiving supplemental oxygen at different timepoints during the study Hour 0 Hour 2 Hour 8 Hour 24 Start of Transfusion End of cerebral tissue O 2 saturation monitor Lactate result used for primary outcome End of lactate monitoring Longer Shorter Longer Shorter Longer Shorter Longer Shorter Room Air Nasal prongs Face mask

7 Cerebral tissue oxygen saturation (to 2 -sat) Cerebral to2-sat was measured non-invasively with a commercially available device (Equanox model 7600, Nonin Corp., Plymouth, MN). Cerebral to2-sat in response to RBC transfusion in one patient is shown in efigure 2. Cerebral readings were taken every four seconds during the 120 minute transfusion using a pediatric near infra-red probe placed on the left forehead. At each 5-minute interval, the mean % saturation from 1 minute (15 readings/minute) was determined and is shown on the y-axis. The total response can be measured as the net area under the curve during the 120 minute transfusion using the % saturation at time 0 as the baseline. The net area under the curve was used to compare outcomes in those receiving shorter-storage vs longer-storage RBCs. See main text. efigure 2. An example of a cerebral tissue oxygen saturation curve

8 eappendix 3. Resolution of impaired consciousness and respiratory distress Changes in the proportion of patients with stupor or coma in each of the two study groups is shown in efigure 3A. Patients continued to improve neurologically following the 24-hour period of observation. Respiratory distress (defined as Kussmaull-type deep breathing, nasal flaring, or inter-costal retractions on inspiration) was common at presentation. Resolution of respiratory distress in each of the two study groups is shown in efigure 3B. efigure 3A. Percentage of participants in each group with stupor or coma observed at different timepoints during the study efigure 3B. Percentage of participants in each group with respiratory distress observed at different timepoints during the study

9 eappendix 4. Vital Signs, arterial O 2 saturation, and electrolytes Serial measurement of the vital signs and arterial O 2 saturation (pulse oximetry) are shown in efigure 4A. Transfusion resulted in rapid improvement in respiratory rate in a time-course compatible with reduction in elevated levels of blood lactate. Plasma bicarbonate, calculated anion gap, potassium, and creatinine are shown in efigure 4B. Values improved with transfusion in each group. Note that transfusion of longerstorage RBCs did not result in elevations of the measured plasma K+. Bars show mean and standard deviation except for arterial O 2 saturation which shows median (lower quartile range). efigure 4A. Vital signs and arterial oxygen saturation among participants in each group measured at different timepoints during the study

10 efigure 4B. Electrolytes and creatinine among participants in each group measured at different timepoints during the study

11 eappendix 5. Adverse events The following adverse events were recorded during the initial 24 hours of observation Longer storage Non-fatal adverse events: unrelated to transfusion Seizures: new onset (not present on admission) 3 4 Seizures present on admission 5 5 Vomiting 4 6 Respiratory distress: new onset 1 0 Non-fatal adverse events: related to transfusion Hives 1 0 Facial edema 0 1 Fatal outcomes during the initial 24 hours Death un-related to transfusion 3 5 Death related to transfusion 0 0 Shorter storage

12 eappendix 6. Subgroup analysis: Patients receiving 2 doses of packed RBCs (20 ml/kg total) Of the 290 enrolled patients, 85 received a second dose of packed RBCs. According to the study protocol, patients were given a second dose of 10 ml/kg of RBCs from the identical unit that was used for the first dose based on the following assessment at hour 4: If the hemoglobin value at hour 4 was <5 g/dl, a second dose was given. If the hemoglobin value at hour 4 was 5-6 g/dl, then a second dose was given if the heart rate was >140 beats per minute (children age 2-5 years) or >160 beats per minute (children age years) and if the respiratory rate at hour 4 was slower than at time 0. Otherwise, a second dose was not given. Among these 85 patients, 47 received longer-storage RBCs and 38 patients received shorter-storage RBCs, p=0.3. Patients who ultimately received a second dose of RBCs were more severely ill on arrival to hospital. Compared with patients treated with a single dose of RBCs, those receiving a second dose were significantly more anemic (2.64 ±0.85 g/dl versus 4.09 ±1.16 g/dl, p<0.0001) and had significantly worse acidosis on arrival as demonstrated by lower plasma CO 2 levels (13.1 ±5.4mM versus 16.6 ±5.23mM, p<0.001), greater elevation of the anion gap (19.8 ±5.4mM versus 17.5 ±4.5mM, p=0.001), and a higher blood lactate level (10.9 ±3.5mM versus 8.7 ±3.1mM, p=0.001). Thus, any impairment of oxygen delivery between the two study groups might be more apparent in this severely ill subgroup. A comparison of response to longer-storage versus shorter-storage RBCs in the 85 patients receiving 20 ml/kg follows. 6.1 Proportion of patients achieving a lactate 3mM at 8 hours: Among those receiving 20 ml/kg of RBCs, the proportion achieving a lactate 3mM at hour 8 was not different between the two groups: 28 of 47 (0.596, 95% CI ) in the longer-storage group and 20 of 38 (0.526, 95% CI ) in the shorter-storage group. The difference in the two groups was 0.07 (95% CI, to 0.27), p= Lactate reduction: Among the subgroup of patients who went on to receive 20 ml/kg of RBCs, mean lactate levels were higher (by chance) at the time of randomization in the longer-storage group (11.7 ±3.2mM versus 9.9 ±3.6mM, p=0.016). However, mean lactate levels were not different between the two groups at hours 2, 4, 6, 8, or 24. See efigure 5A. efigure 5A. Mean blood lactate values in the two study groups among the sub-group of participants transfused with 20 ml/kg of RBCs The error bars indicate the 95% confidence intervals.

13 6.3 Cerebral tissue oxygen saturation Cerebral to 2 -sat significantly increased after transfusion to the same degree in both longer-storage and shorter-storage recipients of 20 ml/kg RBCs. There was no difference in cerebral to 2 -sat between the longer-storage group and the shorter-storage group at the completion of transfusion. See efigure 5B. efigure 5B. Median cerebral tissue oxygen saturation at the start and the end of transfusion for the two study groups among the sub-group of participants transfused with 20 ml/kg of RBCs The horizontal bars indicate the median. The error bars indicate the inter-quartile range (IQR). 6.4 Proportion of patients in stupor or coma at hour 8: The proportion of subjects with persistent stupor or coma at hour 8 was 11 of 47 (0.23) in the longerstorage group and 3 of 38 (0.079) in the shorter-storage group, difference 0.16 (95% CI, to 0.30). 6.5 Proportion of patients who returned to good health at 30 day follow-up: A 30 day telephone follow up of the health status of the patient was available for 59 subjects who received 20 ml/kg of RBCs. The proportion of individuals who had returned to good health was 30 of 36 (0.83) in the longer-storage group and 20 of 23 (0.87) in the shorter-storage group, difference (95% CI, to 0.17). 6.6 Vital signs, electrolytes, BUN, and creatinine in patients receiving 20 ml/kg RBCs.

14 etable 2. Vital signs, electrolytes, BUN and creatinine in the sub-group of participants transfused with 20 ml/kg of RBCs Variable Longer Storage Shorter Storage Difference (95% CI) Result n Result n MAP (torr) Hour ( ) ( ) (-9.23 to 0.10) Hour ( ) ( ) (-4.38 to 4.39) Heart rate Hour ( ) ( (-1.82 to 13.46) 160.6) Hour ( ) ( (1.52 to 16.29) 125.6) Respiratory rate Hour ( ) ( ) (-3.06 to 6.58) Hour ( ) ( ) (-2.37 to 4.58) % O 2 saturation Hour ( ) ( ) (-1.25 to 4.27) Hour ( ) ( ) (-0.37 to 0.69) Potassium (meq/l) Hour ( ) ( ) (0.08 to 0.73) Hour ( ) ( ) (-0.26 to 0.35) CO 2 (mm) Hour ( ) ( ) (-5.17 to -0.37) Hour ( ) ( ) (-2.27 to 0.61) Anion Gap (mm) Hour ( ) ( ) (-2.20 to 2.91) Hour ( ) ( ) (-2.32 to -0.07) BUN (mg/dl) Hour ( ) ( ) ( to 12.19) Hour ( ) ( ) (-9.57 to 8.31) Creatinine (mg/dl) Hour ( ) ( ) (-0.24 to 0.32) Hour ( ) ( ) (-0.17 to 0.26) For each variable, the mean (95% confidence interval) and number tested (n) is shown. CI= confidence interval. MAP=mean arterial pressure

15 eappendix 7. Post-hoc analyses 7.1 Patient factors associated with achieving lactate 3mM by hour 8 In a series of 2x2 association testes, we explored whether or not particular patient factors were associated with the failure to achieve reduction in blood lactate by hour 8. Categories of interest included: age, sex, body mass index, presence of malaria, sickle cell disease, presence of respiratory distress, stupor or coma at presentation, baseline hypoxia, baseline hemoglobin, baseline elevation of B-type natriuretic peptide, mean arterial pressure, pre-transusion intravenous fluids, and ABO type. None of the factors was independently associated with achieving a lactate level 3mM at hour 8. See etable 3 below. etable 3. Patient characteristics among those achieving or not-achieving a lactate of 3 mm or less at hour 8 of the study Factor Lactate 3mM at Lactate >3mM at p-value* hour 8 hour 8 Age <18 months: yes / no 47 / / Baseline MAP<70: yes / no 60 / / Baseline stupor or coma: yes / no 48 / / Baseline respiratory distress: yes / no 149 / / Baseline O 2 saturation <90%: yes / no 5 / / Baseline hemoglobin <3 g/dl: yes / no 58 / / BNP 50pg/mL: yes / no 62 / / Intravenous fluid: yes / no 49 / / BMI <13: yes / no 21 / / Female/male 88 / / Group O/non-Group O 73 / / Malaria: yes / no 138 / / Sickle cell disease: yes / no 22 / / * Fisher exact test 7.2 Malaria patients versus non-malaria patients Of the 290 enrolled patients, 234 had malaria defined as either a positive blood smear or a positive rapid diagnostic test for histidine-rich protein and LDH malaria antigen. Patients with evidence of malaria were treated with anti-malarial medications. We examined baseline levels of lactate and the kinetics of lactate clearance in those with and without malaria to provide evidence in support of the primary study design which analyzed lactate outcomes among all patients regardless of the cause of anemia. Baseline mean lactate levels were identical in those with and without malaria: 9.34 ±3.4mM in malaria patients and 9.34 ±3.4 in non-malaria patients. The proportion of subjects with a lactate 3mM at hour 8 was the same in those with and without malaria: 138 of 232 (59.5%) in those with malaria and 32 of 54 (57.1%) in those without malaria, p=0.99 (Fisher test). Comparison of lactate levels at hours 2, 4, 6, 8, and 24 found no statistically significant differences except at hour 2 where the mean blood lactate was 4.77 ±2.8 in malaria patients and 3.92 ±2.6 in non-malaria patients. No correction was made for multiple comparisons.