ORIGINAL CONTRIBUTION !~l~od,bone marrow aspiration
T
yping and Screening of Blood From
Intraosseous Access
From The Emergency Medicine Residency of St Vincent Medical Center/The Toledo Hospital, Toledo, Ohio.
Kris R Brickman, MD, FACEP Kenneth Krupp, MD Paul Rega, MD, FACEP Jan Alexander, MT (ASCP), SBB Michael Guinness, MD, FACEP
Receivedfor publication January 14, 1991. Revision received November 25, 1991. Accepted for publication December 17, 1991. Presented at the Society
Study objective: Todetermine if intramedullary aspirate from intraosseous needle placement can be used as a source for evaluating blood compatibility.
Design: A prospective, nonrandomized, crossover study. Patients admitted to the hematology/ oncology service undergoing bone marrow aspiration for medical purposes. Setting/participants:
I n t e r v e n t i o n s : Patients had simultaneous samples of bone marrow aspiration from the posterior lilac crest and peripheral venous blood drawn and sent for typing and screening.
.for Academic Emergency
Measurements and main results: The paired samples were
14edicine/Emergency Medicitte
evaluated for ABO and Rh typing as well as the presence of human leukocyte activity by evaluating the reaction strength between the marrow and venous samples.
Research Combined Meeting in Edinburgh, Scotland, October 1990. This study was sponsored in part by FM Douglass Foundation Grant No. 88-80 and by St
Vincent Medical Center.
APRIL 1992 21:4 ANNALS OF EMER6ENCY MEDICINE
Results: No differences were seen in the reaction strength between the paired samples in any subjects for ABO and Rh typing (P= .90, yielding i} = .0523). In addition, human leukocyte activity was detected in both the marrow and venous samples in one patient. Conclusion: Bone marrow aspirates following intraosseous infusion can be used for accurate and reliable typing and screening of blood. [Brickman KR, Krupp K, Rega P, Alexander J, Guinness M: Typing and screening of blood from intraosseous access. Ann EmergMedApril 1992;21:414-417.]
4 1 4/ 8 5
TYPING & SCREENING B r i c k r n a n et al
i
INTRODUCTION Intraosseous (IO) infusion has become a valuable a d j u n c t for the emergency management of critically ill or i n j u r e d children.l-6 This alternative form of vascular access provides a quick and effective route of drug and fluid administration when an IV route is unobtainable. Its use has been restricted p r i m a r i l y to pediatric emergencies such as t r a u ma, burns, cardiac arrest, sudden infant death syndrome, severe dehydration, and anaphylaxis. In t r a u m a patients, this noncollapsible source of vascular access provides a route of fluid resuscitation when a p e r i p h e r a l IV line cannot be obtained due to hypovolemia. Hemorrhagic shock in these patients may require blood transfusion for adequate stabilization. Although blood transfusion has been achieved successfully through the IO route, no studies have evaluated the effectiveness of this transfusion method versus the IV route in pediatric patients. However, the typing and screening of blood from the bone m a r r o w aspirate may provide valuable information in a timely fashion regardless of the mode of transfusion used. Evaluation for blood compatibility could then begin while volume expansion through the IO route is u n d e r way. This study was designed to assess the feasibility of using blood obtained from the bone marrow aspirates in the typing and screening process.
MATERIALS AND METHODS Bone marrow aspirates were obtained from the posterior superior iliae crest of 28 patients who ranged in age from 27 to 84 years (mean, 55.48 years; SD _+54.9) and were admitted to the hematology/ontology service at our institution. F o r each patient, the site was p r e p p e d with povidone-iodine 10% solution, and local anesthesia was obtained with 1% lidocaine. An l l - g a u g e disposable Jamshidi needle was used to aspirate approximately 3 to 5 mL of bone marrow. At the same time, a p e r i p h e r a l venous sample (5 mL) was drawn. This bone marrow aspirate and the venous blood sample then were placed into separate lavender-top tubes containing ethylene diaminetetraacetic acid. All samples were obtained through informed consent. The project and consent form were a p p r o v e d by the Research Committee and the Human Experimentation Committee of St Vincent Medical Center. Bone marrow aspirates and p e r i p h e r a l venous blood samples were tested in parallel for typing and screening according to accepted s t a n d a r d blood b a n k protocol using ABOantisera and D-antisera for ABO front grouping and Rh typing, respectively. A and B reagent RBCs were used for the ABO reverse grouping. Screening reagent RBCs, enhancement solution, and polyspecific antihuman globulin serum were used for the antibody screen. An autocontrol
R 6 /4
1 5
was tested simultaneously with all samples except in two cases in which there was insufficient volume of plasma. All of the above blood typing and antibody screenings were based on visually determined agglutination assays graded on a +1/+2 system. A +2 or above agglutination test would represent a strong, visually significant reaction to the reagent used. Statistical significance of data was evaluated by assessing p r o b a b i l i t y of type II (~) error.
RESULTS In the 28 sample pairs tested, there was no visually significant difference (2+ or more) in the strength of reaction when comparing the m a r r o w aspirate with the p e r i p h e r a l venous samples of each subject (P = .90, yielding ~ = .0523) (Table). In four unrelated sample pairs (two IV and two IO), the ABO reverse grouping and/or the antibody screens could not be compared because of the difficulty in obtaining a sufficient amount of nongelatinous plasma. This problem was due to inadequate sample size and was overcome in a fifth sample by using a filter sampler (Glasrock Products, Inc, F a i r b u r n , Georgia), which forces fibrin-free plasma through the pores of this device and then allows the plasma to be aspirated for compatibility testing. Therefore, the filter sampler was able to provide a larger volume of plasma for typing and screening purposes when the blood sample volume was minimal (less than 3 mL). Two m a r r o w samples showed significant rouleaux formation during the testing process, but this was overcome by saline displacement. No clinically significant red cell alloantibodies were detected in any of the samples. In one subject, cross-reactivity to a previously documented h u m a n •
Table.
Type and screen results comparison Test
Sample Peripheral
Marrow
ABO Type 0
12
12
A
12"
12~
B
2
2
AB
2
2
Rhe (DI positive
21
21
Rho (D) negative
6
6
1
1
27
25~
Rh Type
Rho {D) negative, Du variant positive
Antibody Screen Negative Positive
1t
1
*Plasma quantity insufficient to perform backtype in two cases. tHLA antibody. *Plasma quantity insufficient to perform backtype in b,vo cases not related to above (*} {Plasma quantity insufficient to perform antibody screen in two cases ~lRouleaux displaced by saline in two cases
A N N A L S OF EMERGENCY MEDICINE
21:4
A P R I L 1992
TYPING & SCREENING Brickman et al
leukocyte antibody was noted in both the marrow sample and the p e r i p h e r a l venous blood sample.
DISCUSSION
When pediatric patients present in hemorrhagic shock, blood transfusion may be crucial to their survival. 7 Unfortunately, vascular access for blood transfusion as well as for obtaining samples for typing and screening may be virtually impossible through a p e r i p h e r a l IV route. IO infusion has been proven to be a safe, r a p i d l y established, and effective alternative route of fluid and drug administration when the p e r i p h e r a l IV route is u n o b t a i n a b l e in pediatric patients. 1.4,5,8.9 Recent studies have determined that blood gas measurements and some chemistry values can be measured from the blood aspirate obtained after i n t r a m e d u l l a r y placement in the infusion needle, lO-12 While volume expansion is u n d e r way through the IO route, the blood obtained from the initial IO aspirate could be used to enable the earlier provision of compatible blood. This study was conducted to assess the usefulness of the bone marrow aspirate for determining blood compatibility. This was achieved by comparing bone marrow aspirates with p e r i p h e r a l venous samples drawn simultaneously and sent for typing and screening. The results of this study revealed no clinical difference in the strength of reaction for evaluating blood compatibility between the bone marrow and the p e r i p h e r a l venous sampies. In one adult subject, h u m a n leukocyte antibody was detected in both the marrow and the p e r i p h e r a l venous sample. Although there were no clinically significant red cell alloantibodies detectable in any of the samples, the accurate determination of human leukocyte antibody suggests that alloantibodies detectable in p e r i p h e r a l venous samples by using routine blood b a n k methods also would be detectable in marrow aspirates. Overall, no discrepancies of any type were detected between the p e r i p h e r a l venous and intramedullary samples. In five sample pairs, insufficient amounts of nongelatinous plasma were obtained from both the IV and the bone marrow samples for ABO reverse grouping and antibody screening, but this was overcome by using a filter sampler. Two marrow studies were found to have rouleaux formation d u r ing the testing process, but this was overcome by using saline displacement. Although these samples were obtained in posterior iliac crests from adult patients, bone marrow aspirate from the proximal tibia, which is typically used in pediatric patients, would not be expected to yield additional technical difficulties as long as adequate sample amounts (3 to 5 mL) could be obtained. The type and screening process from each aspiration site in pediatric and adult patients would be identical. In fact, due to the excessively bematogenous marrow in children c o m p a r e d with that in adults, the complica-
APRIL 1992 21:4
ANNALS OF EMERGENCY MEDICINE
tions we encountered (inadequate plasma volume and rouleaux formation) may not be f o u n d in pediatric patients. F u r t h e r studies of a larger sample size in pediatric patients would be i m p o r t a n t to detect any additional problems unique to marrow aspiration from the proximal tibia site in children. Our data revealed that blood a s p i r a t e d from the bone marrow after IO placement can be used for accurate typing and screening of blood. Although blood transfusions have previously been given through the IO site, the effectiveness of this site compared with the IV route has yet to be established in pediatric patients. 1,13-16 Complications p r i m a r i l y involve inadequate sample volume and can be overcome with additional equipment (ie, filter sampler to attain higher plasma volume or saline dilution to compensate for excessive rouleaux formation).
CONCLUSION
Our d a t a suggest that blood a s p i r a t e d from the bone marrow after IO placement can be used reliably for the typing and screening of blood in adults. Such aspirates may be useful in the expeditious provision of compatible blood for transfusion when blood samples are initially unobtainable from an IV site. Volume expansion through IO infusion then can be followed by a more easily placed p e r i p h e r a l IV line and transfusion of compatible blood products as necessary. Due to problems of rouleaux formation or inadequate aspirate sample amount, saline displacement of the sample or a special filter may be needed for complete typing and screening. • The authors t h a n k B a r b a r a Blank, CMT, for p r e p a r a t i o n of the manuscript. REFERENCES 1. Rosetti VA, Thompson BM, Miller J, et al: Intraosseous infusion: An alternative route of pediatric intravascular access. Ann Emerg Med 1985;14:885-888. 2. Odowski JP: My kingdom for an intravenous line. AmJDis Child1984;138:803. 3. Kanter RK, Zimmerman J J, Strauss RH, et al: Pediatric emergency intravenous access. Am J Dis Child 1988;140:132-134. 4. Hodge: Intraosseous infusion: A review. PediatrEmerg Care 1988;1:215-218. 5. Smith RJ, Keseg DP, Manley LK, et al: fntraosseous infusion by prehospital personnel in critically ill pediatric patient. Ann Emerg Med 1988;17:491-495. 6. Brickman KR, Rega P, Guinness M: Intraosseous infusion: Rapid vascular access in critically ill children. Emerg Med Serv 1989;18:56-58. 7. Shires GT, Canizaro PC: Fluid resuscitation in the severely injured child. Surg Clin North Am 1973;53:1341. 8. Berg RA: Emergency infusion of catecholamine into bone marrow. Am J Dis Child 1986;138:810-811. 9, Valdes MM: Intraosseous fluid administration in emergencies. Lancet1977;1:12351236. •
416/8
7
TYPING & SCREENING B r i c k m a n et al
10. Brickman KR: A comparative study of intraosseous, intravenous and intra-arterial pH changes during hypoventilation in dogs (abstract). Ann Emerg Med 1987;18:183.
Address for reprints: Kris R Brickman, MD, FACEP, 5305 River Oaks Court, Sylvania Ohio 43560.
11. Spivey WH, McNamara RM, MacKenzie RS: Comparison of intraosseous and intravenous CBC and ASTRA8 in swine (abstract). Ann Emerg Med 1986;15:647. 12. Orlowski JT, Porembka DT, Oallagher JM, et al: The bone marrow as a source of laboratory studies. Ann Emerg Med 1989;18:1348-1351. 13. Tocantins LM, O'Neill JF: Infusions of blood and other fluids into the general circulation via the bone marrow. Surg Gynecol Obstet 1941;73:28t-287. 14. Tacantins LM, O'Neill JF, Price AH: Infusion of blood and other fluids via the bone marrow in traumatic shock and other forms of peripheral circulatory failure. Ann Surg 1941;114:1085-1092. 15. Bailey H: Bone marrow as a site for the reception of infusions, transfusions, and anesthetic agents. Br Med J (Clin Res) 1944;2:181-182. 16. Henning N: Intrasternal injections and transfusions. JAMA 1945;128:240.
88/417
ANNALS OF EMERGENCY MEDICINE
21:4
APRIL1992
T
yping and Screening of Blood From
Intraosseous Access
From The Emergency Medicine Residency of St Vincent Medical Center/The Toledo Hospital, Toledo, Ohio.
Kris R Brickman, MD, FACEP Kenneth Krupp, MD Paul Rega, MD, FACEP Jan Alexander, MT (ASCP), SBB Michael Guinness, MD, FACEP
Receivedfor publication January 14, 1991. Revision received November 25, 1991. Accepted for publication December 17, 1991. Presented at the Society
Study objective: Todetermine if intramedullary aspirate from intraosseous needle placement can be used as a source for evaluating blood compatibility.
Design: A prospective, nonrandomized, crossover study. Patients admitted to the hematology/ oncology service undergoing bone marrow aspiration for medical purposes. Setting/participants:
I n t e r v e n t i o n s : Patients had simultaneous samples of bone marrow aspiration from the posterior lilac crest and peripheral venous blood drawn and sent for typing and screening.
.for Academic Emergency
Measurements and main results: The paired samples were
14edicine/Emergency Medicitte
evaluated for ABO and Rh typing as well as the presence of human leukocyte activity by evaluating the reaction strength between the marrow and venous samples.
Research Combined Meeting in Edinburgh, Scotland, October 1990. This study was sponsored in part by FM Douglass Foundation Grant No. 88-80 and by St
Vincent Medical Center.
APRIL 1992 21:4 ANNALS OF EMER6ENCY MEDICINE
Results: No differences were seen in the reaction strength between the paired samples in any subjects for ABO and Rh typing (P= .90, yielding i} = .0523). In addition, human leukocyte activity was detected in both the marrow and venous samples in one patient. Conclusion: Bone marrow aspirates following intraosseous infusion can be used for accurate and reliable typing and screening of blood. [Brickman KR, Krupp K, Rega P, Alexander J, Guinness M: Typing and screening of blood from intraosseous access. Ann EmergMedApril 1992;21:414-417.]
4 1 4/ 8 5
TYPING & SCREENING B r i c k r n a n et al
i
INTRODUCTION Intraosseous (IO) infusion has become a valuable a d j u n c t for the emergency management of critically ill or i n j u r e d children.l-6 This alternative form of vascular access provides a quick and effective route of drug and fluid administration when an IV route is unobtainable. Its use has been restricted p r i m a r i l y to pediatric emergencies such as t r a u ma, burns, cardiac arrest, sudden infant death syndrome, severe dehydration, and anaphylaxis. In t r a u m a patients, this noncollapsible source of vascular access provides a route of fluid resuscitation when a p e r i p h e r a l IV line cannot be obtained due to hypovolemia. Hemorrhagic shock in these patients may require blood transfusion for adequate stabilization. Although blood transfusion has been achieved successfully through the IO route, no studies have evaluated the effectiveness of this transfusion method versus the IV route in pediatric patients. However, the typing and screening of blood from the bone m a r r o w aspirate may provide valuable information in a timely fashion regardless of the mode of transfusion used. Evaluation for blood compatibility could then begin while volume expansion through the IO route is u n d e r way. This study was designed to assess the feasibility of using blood obtained from the bone marrow aspirates in the typing and screening process.
MATERIALS AND METHODS Bone marrow aspirates were obtained from the posterior superior iliae crest of 28 patients who ranged in age from 27 to 84 years (mean, 55.48 years; SD _+54.9) and were admitted to the hematology/ontology service at our institution. F o r each patient, the site was p r e p p e d with povidone-iodine 10% solution, and local anesthesia was obtained with 1% lidocaine. An l l - g a u g e disposable Jamshidi needle was used to aspirate approximately 3 to 5 mL of bone marrow. At the same time, a p e r i p h e r a l venous sample (5 mL) was drawn. This bone marrow aspirate and the venous blood sample then were placed into separate lavender-top tubes containing ethylene diaminetetraacetic acid. All samples were obtained through informed consent. The project and consent form were a p p r o v e d by the Research Committee and the Human Experimentation Committee of St Vincent Medical Center. Bone marrow aspirates and p e r i p h e r a l venous blood samples were tested in parallel for typing and screening according to accepted s t a n d a r d blood b a n k protocol using ABOantisera and D-antisera for ABO front grouping and Rh typing, respectively. A and B reagent RBCs were used for the ABO reverse grouping. Screening reagent RBCs, enhancement solution, and polyspecific antihuman globulin serum were used for the antibody screen. An autocontrol
R 6 /4
1 5
was tested simultaneously with all samples except in two cases in which there was insufficient volume of plasma. All of the above blood typing and antibody screenings were based on visually determined agglutination assays graded on a +1/+2 system. A +2 or above agglutination test would represent a strong, visually significant reaction to the reagent used. Statistical significance of data was evaluated by assessing p r o b a b i l i t y of type II (~) error.
RESULTS In the 28 sample pairs tested, there was no visually significant difference (2+ or more) in the strength of reaction when comparing the m a r r o w aspirate with the p e r i p h e r a l venous samples of each subject (P = .90, yielding ~ = .0523) (Table). In four unrelated sample pairs (two IV and two IO), the ABO reverse grouping and/or the antibody screens could not be compared because of the difficulty in obtaining a sufficient amount of nongelatinous plasma. This problem was due to inadequate sample size and was overcome in a fifth sample by using a filter sampler (Glasrock Products, Inc, F a i r b u r n , Georgia), which forces fibrin-free plasma through the pores of this device and then allows the plasma to be aspirated for compatibility testing. Therefore, the filter sampler was able to provide a larger volume of plasma for typing and screening purposes when the blood sample volume was minimal (less than 3 mL). Two m a r r o w samples showed significant rouleaux formation during the testing process, but this was overcome by saline displacement. No clinically significant red cell alloantibodies were detected in any of the samples. In one subject, cross-reactivity to a previously documented h u m a n •
Table.
Type and screen results comparison Test
Sample Peripheral
Marrow
ABO Type 0
12
12
A
12"
12~
B
2
2
AB
2
2
Rhe (DI positive
21
21
Rho (D) negative
6
6
1
1
27
25~
Rh Type
Rho {D) negative, Du variant positive
Antibody Screen Negative Positive
1t
1
*Plasma quantity insufficient to perform backtype in two cases. tHLA antibody. *Plasma quantity insufficient to perform backtype in b,vo cases not related to above (*} {Plasma quantity insufficient to perform antibody screen in two cases ~lRouleaux displaced by saline in two cases
A N N A L S OF EMERGENCY MEDICINE
21:4
A P R I L 1992
TYPING & SCREENING Brickman et al
leukocyte antibody was noted in both the marrow sample and the p e r i p h e r a l venous blood sample.
DISCUSSION
When pediatric patients present in hemorrhagic shock, blood transfusion may be crucial to their survival. 7 Unfortunately, vascular access for blood transfusion as well as for obtaining samples for typing and screening may be virtually impossible through a p e r i p h e r a l IV route. IO infusion has been proven to be a safe, r a p i d l y established, and effective alternative route of fluid and drug administration when the p e r i p h e r a l IV route is u n o b t a i n a b l e in pediatric patients. 1.4,5,8.9 Recent studies have determined that blood gas measurements and some chemistry values can be measured from the blood aspirate obtained after i n t r a m e d u l l a r y placement in the infusion needle, lO-12 While volume expansion is u n d e r way through the IO route, the blood obtained from the initial IO aspirate could be used to enable the earlier provision of compatible blood. This study was conducted to assess the usefulness of the bone marrow aspirate for determining blood compatibility. This was achieved by comparing bone marrow aspirates with p e r i p h e r a l venous samples drawn simultaneously and sent for typing and screening. The results of this study revealed no clinical difference in the strength of reaction for evaluating blood compatibility between the bone marrow and the p e r i p h e r a l venous sampies. In one adult subject, h u m a n leukocyte antibody was detected in both the marrow and the p e r i p h e r a l venous sample. Although there were no clinically significant red cell alloantibodies detectable in any of the samples, the accurate determination of human leukocyte antibody suggests that alloantibodies detectable in p e r i p h e r a l venous samples by using routine blood b a n k methods also would be detectable in marrow aspirates. Overall, no discrepancies of any type were detected between the p e r i p h e r a l venous and intramedullary samples. In five sample pairs, insufficient amounts of nongelatinous plasma were obtained from both the IV and the bone marrow samples for ABO reverse grouping and antibody screening, but this was overcome by using a filter sampler. Two marrow studies were found to have rouleaux formation d u r ing the testing process, but this was overcome by using saline displacement. Although these samples were obtained in posterior iliac crests from adult patients, bone marrow aspirate from the proximal tibia, which is typically used in pediatric patients, would not be expected to yield additional technical difficulties as long as adequate sample amounts (3 to 5 mL) could be obtained. The type and screening process from each aspiration site in pediatric and adult patients would be identical. In fact, due to the excessively bematogenous marrow in children c o m p a r e d with that in adults, the complica-
APRIL 1992 21:4
ANNALS OF EMERGENCY MEDICINE
tions we encountered (inadequate plasma volume and rouleaux formation) may not be f o u n d in pediatric patients. F u r t h e r studies of a larger sample size in pediatric patients would be i m p o r t a n t to detect any additional problems unique to marrow aspiration from the proximal tibia site in children. Our data revealed that blood a s p i r a t e d from the bone marrow after IO placement can be used for accurate typing and screening of blood. Although blood transfusions have previously been given through the IO site, the effectiveness of this site compared with the IV route has yet to be established in pediatric patients. 1,13-16 Complications p r i m a r i l y involve inadequate sample volume and can be overcome with additional equipment (ie, filter sampler to attain higher plasma volume or saline dilution to compensate for excessive rouleaux formation).
CONCLUSION
Our d a t a suggest that blood a s p i r a t e d from the bone marrow after IO placement can be used reliably for the typing and screening of blood in adults. Such aspirates may be useful in the expeditious provision of compatible blood for transfusion when blood samples are initially unobtainable from an IV site. Volume expansion through IO infusion then can be followed by a more easily placed p e r i p h e r a l IV line and transfusion of compatible blood products as necessary. Due to problems of rouleaux formation or inadequate aspirate sample amount, saline displacement of the sample or a special filter may be needed for complete typing and screening. • The authors t h a n k B a r b a r a Blank, CMT, for p r e p a r a t i o n of the manuscript. REFERENCES 1. Rosetti VA, Thompson BM, Miller J, et al: Intraosseous infusion: An alternative route of pediatric intravascular access. Ann Emerg Med 1985;14:885-888. 2. Odowski JP: My kingdom for an intravenous line. AmJDis Child1984;138:803. 3. Kanter RK, Zimmerman J J, Strauss RH, et al: Pediatric emergency intravenous access. Am J Dis Child 1988;140:132-134. 4. Hodge: Intraosseous infusion: A review. PediatrEmerg Care 1988;1:215-218. 5. Smith RJ, Keseg DP, Manley LK, et al: fntraosseous infusion by prehospital personnel in critically ill pediatric patient. Ann Emerg Med 1988;17:491-495. 6. Brickman KR, Rega P, Guinness M: Intraosseous infusion: Rapid vascular access in critically ill children. Emerg Med Serv 1989;18:56-58. 7. Shires GT, Canizaro PC: Fluid resuscitation in the severely injured child. Surg Clin North Am 1973;53:1341. 8. Berg RA: Emergency infusion of catecholamine into bone marrow. Am J Dis Child 1986;138:810-811. 9, Valdes MM: Intraosseous fluid administration in emergencies. Lancet1977;1:12351236. •
416/8
7
TYPING & SCREENING B r i c k m a n et al
10. Brickman KR: A comparative study of intraosseous, intravenous and intra-arterial pH changes during hypoventilation in dogs (abstract). Ann Emerg Med 1987;18:183.
Address for reprints: Kris R Brickman, MD, FACEP, 5305 River Oaks Court, Sylvania Ohio 43560.
11. Spivey WH, McNamara RM, MacKenzie RS: Comparison of intraosseous and intravenous CBC and ASTRA8 in swine (abstract). Ann Emerg Med 1986;15:647. 12. Orlowski JT, Porembka DT, Oallagher JM, et al: The bone marrow as a source of laboratory studies. Ann Emerg Med 1989;18:1348-1351. 13. Tocantins LM, O'Neill JF: Infusions of blood and other fluids into the general circulation via the bone marrow. Surg Gynecol Obstet 1941;73:28t-287. 14. Tacantins LM, O'Neill JF, Price AH: Infusion of blood and other fluids via the bone marrow in traumatic shock and other forms of peripheral circulatory failure. Ann Surg 1941;114:1085-1092. 15. Bailey H: Bone marrow as a site for the reception of infusions, transfusions, and anesthetic agents. Br Med J (Clin Res) 1944;2:181-182. 16. Henning N: Intrasternal injections and transfusions. JAMA 1945;128:240.
88/417
ANNALS OF EMERGENCY MEDICINE
21:4
APRIL1992
