Blood flow in muscle groups and drug absorption Resting human muscle blood flow (MBF) was determined simultaneously in the usual intramuscular injection sites to resolve whether variance in MBF could account for differences in drug absorption. Three pairs of muscles (gluteus maximus, vastus lateralis, and deltoid) were studied in each of 20 adult subjects. Use of dual, matched linear rate meters allowed two muscles to be studied simultaneously, with the order of injection random within an incomplete block design. MBF was calculated from the 133xenon washout rate using a single exponential that the computer found to best fit the data. Deltoid MBF (11.6 mll 100 gmlmin ± 0.5) was significantly (p < 0.05) greater than gluteus MBF (9.6 ± 0.5), with vastus MBF intermediate (10.8 ± 0.6). There were no significant differences due to order of injection or between right and left sides for each muscle. These data indicate that there are consistent differences in resting MBF among specific muscle groups of sufficient magnitude (19%) to affect the rate of absorption and peak serum levels following intramuscular administration of drugs.
Eleanor F. Evans, M.D., Jack D. Proctor, M.D., Melvin J. Fratkin, M.D., Jorge Velandia, M.D., and Albert J. Wasserman, M.D. Richmond, Va. Division of Clinical Pharmacology, Department of Medicine and Department of Radiology, Medical College of Virginia
Little attention has been directed to differences between muscle sites into which injections of drugs are made. Recent reports present evidence that for several drugs blood levels differ with muscles used for injection. Binder1 found that intramuscular insulin absorption correlated to regional blood flow. Cohen and asSupported in part by a grant from the Small Grants Committee. School of Medicine, Medical College of Virginia, Virginia Commonwealth University. Presented in part at the Seventy-fourth Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics. New Orleans. La., March 23, 1973. Received for publication March 26, 1974. Accepted for publication Oct. 7, 1974. Reprint requests to: Eleanor F. Evans, M.O., Box 284, Medical College of Virginia, Richmond. Va. 23298.
44
sociates 4 and Schwartz and associates ll demonstrated that the site of intramuscular injection of lidocaine significantly influenced achievable plasma levels. In a similar study, 12 gamma globulin uptake and blood levels differed according to the site of intramuscular injection. Measurement of muscle blood flow in normal subjects has been accomplished successfully by means of 133xenon (133Xe) washout. 3, 8, 9 Comparative resting muscle blood flow in muscles commonly used for intramuscular injection has not been systematically studied. To investigate the possible differences in muscle blood flow, 133Xe washout from the deltoid, vastus lateralis, and gluteal muscles was measured in normal subjects in a resting state.
Blood flow in muscle and absorption
Volume 17 Number I
60
----
RIGHT GLUTEUS
RIGHT DEL TOlD
40 20
CPM xl0 3
45
0 LEFT GLUTEUS
60
40
" '. ~ . '.
20
'"
2
4
8
6
o
2
4
6
8
TIME (minutes)
Fig. 1. 133Xenon disappearance curves obtained from Subject 02. Differences between the muscle groups may be appreciated. In this subject, opposite members of the same muscle group were injected simultaneously. The atypical curve for the right gluteus muscle is explained in the text. The straight lines were fitted from the computer-calculated slopes using the data from the entire 8 minutes. Methods
The indicator washout method using radioactive xenon selected for measurement of muscle blood flow was validated by Lassen, Lindbjerg, and Munck. 9 It was assumed by us, as by Lassen, that the diffusion equilibrium of Xe between muscle and blood was maintained under these conditions. The equation of Lassen9 was used for all calculations. The method requires only the injection of 133Xe into the muscle and the measurement of the rate at which it is removed by the blood as indicated by its disappearance from the injection site. This is easily done by continuous counting over the site by external scintillation detectors. Xenon is an inert gas that, because it is lipophilic, freely diffuses through muscle cell membranes to blood with no compartment to trap the tracer. 9 Disappearance of the isotope is, therefore, flow-limited, and local blood flow can be quantitatively ascertained. There is little or no 133Xe recirculation because it is rapidly cleared from the blood in first passage through the lungs and is excreted in expired air. 2, 10 Twenty healthy men between the ages of 21 and 30 served as subjects. All subjects were studied while recumbent and apparently relaxed. Expired air containing excreted 133Xe was exhausted from the area of the detectors. About 50 /Lei of 133Xe dissolved in sterile isotonic saline solution in 0.2 ml was injected into
133XENON MUSCLE BLOOD FLOW RESTING HUMAN
D> G
D >V V> G
12
P(O.5 N.S. N.S.
10
t::
8
~
"E:
t),
6
CJ
2
"..... E:
4
2
0
39
39
35
DELTOID
VASTUS
GLUTEUS
IL6±05(SEM)
108"06
9.6 ± 0.5
Fig. 2. Mean blood flows for each muscle group are compared. The differences are significant only between deltoid and gluteus muscles.
each of the paired muscle sites. A 26-ga uge 11;2inch disposable needle was used for injections. Great care was taken to minimize trauma. Two sites were injected simultaneously and the disappearance rate of the isotope was measured with matched sodium iodide scintillation detectors connected through rate meters to re-
46
Evans et ai.
cording potentiometers. The paired injection sites for each subject were randomized within an incomplete block design. Counting began approximately 45 sec after initiation of the injections and was continued for 8 min. All three pairs of injection sites were used in every subject, a total of 120 muscle sites. Fig. 1 shows the l33Xe disappearance curves in Subject 02. The flat curve from the right gluteus muscle was thought to depict accidental injection into the fatty tissue above the gluteal muscle. This very slow clearance may represent both the low level of fat blood flow and high affinity of Xe for fat and thus little clearance from the site during observation. This curve and 4 such curves in other subjects were eliminated from analysis. The numerical counts were taken from the direct writing potentiometer plot at 30-sec intervals for 8 min, using a time constant of 3 sec and a paper speed of 0.2 inch per min, and the data were entered into a computer. Muscle blood flow for each site was determined according to the equation: F( mIl 100 gm/ min) = -70 x In( lO)d loglOQ/ dt = 161 x d loglo Q/dt
d loglo Q/dt is obtained from the slope of the least squares straight line of loglo of the count rate Q vs time. Here F is the muscle blood flow in ml of blood per min per 100 gm of tissue and Q is the count rate at the time t min after injection. All muscle blood flow determinations were subjected to analysis of variance, testing also effects due to order of injection and the side of injection site. Differences in muscle blood flow were sought. No differences were realized due to order of injection or between right and left sides of the body. There were differences between muscles. Only those between the deltoid and the gluteus muscles were significant when tested by Duncan's Multiple Range Test. l3 The mean blood flow to the deltoid muscle was 11.6 ml/ 100 gm/min and mean blood flow to the gluteus muscle was 9.6 ml/IOO gm/min, a difference of 19% (p < 0.05). Flow to the vastus lateralis muscle was intermediate, being 10.8 ml/100
Clinical Pharmacology and Therapeutics
gm/ min, and was not significantl y different from either the deltoid or gluteus muscle (Fig. 2). Discussion
To our knowledge, this study represents the only systematic investigation of differences in peak muscle blood flow measured simultaneously in the same individual. Our results for resting muscle blood flow seem to differ from similar reports found in the literature. The calculated muscle blood flow reported by Lassen, Lindbjerg, and Munck9 and Corman and associates,6 using the tibialis anterior muscle, was 2 ml/ 100 gm/min, a value approximately one-fifth the 10 to 11 ml/100 gm/min we found in the vastus or deltoid muscles. The difference is easi1y explainable on the basis of the part of the disappearance slope used to calculate muscle blood flow. The initial portion of this slope was included in our study and, being the steepest component of the washout, it produced a higher calculated flow than that of Corman and Lassen who used only the second or leveling portion. The initial steep portion was the proper portion to use because the disappearance of xenon from each muscle was the point of the investigation. Binderl also noted this problem and in his work referred to the early disappearance as the T -50, namely, the time from injection until only 50% of the injected drug remained at the injection site. Although our blood flow results were higher than those previously reported, the conclusion that xenon disappearance was significantly faster from the deltoid than from the gluteus maximus is valid. The two sites most commonly used for intramuscular injections are the deltoid and the gluteus muscles. The importance of the difference we found as a determinant of peak drug blood levels and, therefore, of rate of development of drug action is implied. However, further study is needed to determine for which drugs the difference has clinical significance. Lidocaine is one drug that has been investigated for the effect of injection site. 4 , 11 Deltoid injection gave higher peak levels than lateral thigh injection, which in tum gave higher levels than gluteal injection. Therapeutic plasma levels for arrhythmia prevention with doses of
Volume 17 Number J
4.5 mg/kg of a 10% solution were reached only when the deltoid injection site was used. l l This demonstrates that the site of injection can influence the plasma level achieved and that the deltoid muscle should be used to achieve therapeutic blood levels as rapidly as possible. We wish to thank Dr. Roger Flora of the Department of Biometry for the computer programming and statistical analyses, Dr. James Worsham for consultations in theory and techniques, and Mrs. Robert T. Dance, Jr., for her technical assistance.
Blood flow in muscle and absorption
6.
7. 8.
References I. Binder, c.: Absorption of injected insulin, thesis, Hvidore Hospital and Novo Research Institute, Copenhagen, 1969, Ejnar Munksgaards Forlag. 2. Chidsey, C. A., Fritts, H. W., Jr., Hardewig, A., Richards, D. W., and Cournand, A.: Fate of radioactive krypton (Kr 85 ) introduced intravenously in man, J. App!. Physiol. 14:63-66, 1959. 3. Clausen, J. P., and Lassen, N. A.: Muscle blood flow during exercise in normal man studied by the 133Xenon clearance method, Cardiovasc. Res. 5:245-254, 1971. 4. Cohen, L. S., Rosenthal, J. E., Horner, D. W., Jr., Atkins, J. M., Matthews, O. A., and Sarnoff, S. J.: Plasma levels of lidocaine after intramuscular administration, Am. J. Cardio!. 29:520-523, 1972. 5. Conn, H. L., Jr.: Equilibrium distribution of radio xenon in tissue: Xenon-hemoglobin associ-
9.
10.
11.
12.
13.
47
ation curve, J. App!. Physio!. 16: 1065- 1070, 1961. Corman, L. A., Flicklinger, F. W., Sokoloff, J., and Nodine, J. H.: Radioactive Xenon tissue clearance: Standardization for measurement of peripheral blood flow, J. Nuc!. Med. 11:233238, 1970. Goldstein, A.: Biostatistics, an introductory text. New York, 1964, The Macmillan Company, p. 136. Holzman, G. B., Wagner, H. N., Iio, M., Rabinowitz, D., and Zierler, K. L.: Measurement of muscle blood flow in the human forearm with radioactive krypton and xenon, Circulation 30:27-34, 1964. Lassen, N. A., Lindbjerg, J., and Munck, 0.: Measurement of blood flow through skeletal muscle by intramuscular injection of Xenon-133, Lancet 1:686-689, 1964. Loken, M. K., and Westgate, H. D.: Using Xenon-133 and a scintillation camera to evaluate pulmonary function, J. Nucl. Med. 9:45-50, 1967. Schwartz, M. L., Meyer, M. B., Covino, B. G., Narang, R. M., Sethi, Y., Schwartz, A. J., and Kamp, P.: Antiarrhythmic effectiveness of intramuscular lidocaine; influence of different injection sites, J. Clin. Pharmacol. 14:77-83, 1974. Smith, G. N., Mollison, D., Griffiths, B., and Mollison, P. L.: Uptake ofigG after intramuscular and subcutaneous injection, Lancet 1: 12081212, 1972. Steel, R. G. D., and Torrie, J. H.: Principles and procedures of statistics, New York, 1960, McGraw-Hill Book Co., p. 107.
Eleanor F. Evans, M.D., Jack D. Proctor, M.D., Melvin J. Fratkin, M.D., Jorge Velandia, M.D., and Albert J. Wasserman, M.D. Richmond, Va. Division of Clinical Pharmacology, Department of Medicine and Department of Radiology, Medical College of Virginia
Little attention has been directed to differences between muscle sites into which injections of drugs are made. Recent reports present evidence that for several drugs blood levels differ with muscles used for injection. Binder1 found that intramuscular insulin absorption correlated to regional blood flow. Cohen and asSupported in part by a grant from the Small Grants Committee. School of Medicine, Medical College of Virginia, Virginia Commonwealth University. Presented in part at the Seventy-fourth Annual Meeting of the American Society for Clinical Pharmacology and Therapeutics. New Orleans. La., March 23, 1973. Received for publication March 26, 1974. Accepted for publication Oct. 7, 1974. Reprint requests to: Eleanor F. Evans, M.O., Box 284, Medical College of Virginia, Richmond. Va. 23298.
44
sociates 4 and Schwartz and associates ll demonstrated that the site of intramuscular injection of lidocaine significantly influenced achievable plasma levels. In a similar study, 12 gamma globulin uptake and blood levels differed according to the site of intramuscular injection. Measurement of muscle blood flow in normal subjects has been accomplished successfully by means of 133xenon (133Xe) washout. 3, 8, 9 Comparative resting muscle blood flow in muscles commonly used for intramuscular injection has not been systematically studied. To investigate the possible differences in muscle blood flow, 133Xe washout from the deltoid, vastus lateralis, and gluteal muscles was measured in normal subjects in a resting state.
Blood flow in muscle and absorption
Volume 17 Number I
60
----
RIGHT GLUTEUS
RIGHT DEL TOlD
40 20
CPM xl0 3
45
0 LEFT GLUTEUS
60
40
" '. ~ . '.
20
'"
2
4
8
6
o
2
4
6
8
TIME (minutes)
Fig. 1. 133Xenon disappearance curves obtained from Subject 02. Differences between the muscle groups may be appreciated. In this subject, opposite members of the same muscle group were injected simultaneously. The atypical curve for the right gluteus muscle is explained in the text. The straight lines were fitted from the computer-calculated slopes using the data from the entire 8 minutes. Methods
The indicator washout method using radioactive xenon selected for measurement of muscle blood flow was validated by Lassen, Lindbjerg, and Munck. 9 It was assumed by us, as by Lassen, that the diffusion equilibrium of Xe between muscle and blood was maintained under these conditions. The equation of Lassen9 was used for all calculations. The method requires only the injection of 133Xe into the muscle and the measurement of the rate at which it is removed by the blood as indicated by its disappearance from the injection site. This is easily done by continuous counting over the site by external scintillation detectors. Xenon is an inert gas that, because it is lipophilic, freely diffuses through muscle cell membranes to blood with no compartment to trap the tracer. 9 Disappearance of the isotope is, therefore, flow-limited, and local blood flow can be quantitatively ascertained. There is little or no 133Xe recirculation because it is rapidly cleared from the blood in first passage through the lungs and is excreted in expired air. 2, 10 Twenty healthy men between the ages of 21 and 30 served as subjects. All subjects were studied while recumbent and apparently relaxed. Expired air containing excreted 133Xe was exhausted from the area of the detectors. About 50 /Lei of 133Xe dissolved in sterile isotonic saline solution in 0.2 ml was injected into
133XENON MUSCLE BLOOD FLOW RESTING HUMAN
D> G
D >V V> G
12
P(O.5 N.S. N.S.
10
t::
8
~
"E:
t),
6
CJ
2
"..... E:
4
2
0
39
39
35
DELTOID
VASTUS
GLUTEUS
IL6±05(SEM)
108"06
9.6 ± 0.5
Fig. 2. Mean blood flows for each muscle group are compared. The differences are significant only between deltoid and gluteus muscles.
each of the paired muscle sites. A 26-ga uge 11;2inch disposable needle was used for injections. Great care was taken to minimize trauma. Two sites were injected simultaneously and the disappearance rate of the isotope was measured with matched sodium iodide scintillation detectors connected through rate meters to re-
46
Evans et ai.
cording potentiometers. The paired injection sites for each subject were randomized within an incomplete block design. Counting began approximately 45 sec after initiation of the injections and was continued for 8 min. All three pairs of injection sites were used in every subject, a total of 120 muscle sites. Fig. 1 shows the l33Xe disappearance curves in Subject 02. The flat curve from the right gluteus muscle was thought to depict accidental injection into the fatty tissue above the gluteal muscle. This very slow clearance may represent both the low level of fat blood flow and high affinity of Xe for fat and thus little clearance from the site during observation. This curve and 4 such curves in other subjects were eliminated from analysis. The numerical counts were taken from the direct writing potentiometer plot at 30-sec intervals for 8 min, using a time constant of 3 sec and a paper speed of 0.2 inch per min, and the data were entered into a computer. Muscle blood flow for each site was determined according to the equation: F( mIl 100 gm/ min) = -70 x In( lO)d loglOQ/ dt = 161 x d loglo Q/dt
d loglo Q/dt is obtained from the slope of the least squares straight line of loglo of the count rate Q vs time. Here F is the muscle blood flow in ml of blood per min per 100 gm of tissue and Q is the count rate at the time t min after injection. All muscle blood flow determinations were subjected to analysis of variance, testing also effects due to order of injection and the side of injection site. Differences in muscle blood flow were sought. No differences were realized due to order of injection or between right and left sides of the body. There were differences between muscles. Only those between the deltoid and the gluteus muscles were significant when tested by Duncan's Multiple Range Test. l3 The mean blood flow to the deltoid muscle was 11.6 ml/ 100 gm/min and mean blood flow to the gluteus muscle was 9.6 ml/IOO gm/min, a difference of 19% (p < 0.05). Flow to the vastus lateralis muscle was intermediate, being 10.8 ml/100
Clinical Pharmacology and Therapeutics
gm/ min, and was not significantl y different from either the deltoid or gluteus muscle (Fig. 2). Discussion
To our knowledge, this study represents the only systematic investigation of differences in peak muscle blood flow measured simultaneously in the same individual. Our results for resting muscle blood flow seem to differ from similar reports found in the literature. The calculated muscle blood flow reported by Lassen, Lindbjerg, and Munck9 and Corman and associates,6 using the tibialis anterior muscle, was 2 ml/ 100 gm/min, a value approximately one-fifth the 10 to 11 ml/100 gm/min we found in the vastus or deltoid muscles. The difference is easi1y explainable on the basis of the part of the disappearance slope used to calculate muscle blood flow. The initial portion of this slope was included in our study and, being the steepest component of the washout, it produced a higher calculated flow than that of Corman and Lassen who used only the second or leveling portion. The initial steep portion was the proper portion to use because the disappearance of xenon from each muscle was the point of the investigation. Binderl also noted this problem and in his work referred to the early disappearance as the T -50, namely, the time from injection until only 50% of the injected drug remained at the injection site. Although our blood flow results were higher than those previously reported, the conclusion that xenon disappearance was significantly faster from the deltoid than from the gluteus maximus is valid. The two sites most commonly used for intramuscular injections are the deltoid and the gluteus muscles. The importance of the difference we found as a determinant of peak drug blood levels and, therefore, of rate of development of drug action is implied. However, further study is needed to determine for which drugs the difference has clinical significance. Lidocaine is one drug that has been investigated for the effect of injection site. 4 , 11 Deltoid injection gave higher peak levels than lateral thigh injection, which in tum gave higher levels than gluteal injection. Therapeutic plasma levels for arrhythmia prevention with doses of
Volume 17 Number J
4.5 mg/kg of a 10% solution were reached only when the deltoid injection site was used. l l This demonstrates that the site of injection can influence the plasma level achieved and that the deltoid muscle should be used to achieve therapeutic blood levels as rapidly as possible. We wish to thank Dr. Roger Flora of the Department of Biometry for the computer programming and statistical analyses, Dr. James Worsham for consultations in theory and techniques, and Mrs. Robert T. Dance, Jr., for her technical assistance.
Blood flow in muscle and absorption
6.
7. 8.
References I. Binder, c.: Absorption of injected insulin, thesis, Hvidore Hospital and Novo Research Institute, Copenhagen, 1969, Ejnar Munksgaards Forlag. 2. Chidsey, C. A., Fritts, H. W., Jr., Hardewig, A., Richards, D. W., and Cournand, A.: Fate of radioactive krypton (Kr 85 ) introduced intravenously in man, J. App!. Physiol. 14:63-66, 1959. 3. Clausen, J. P., and Lassen, N. A.: Muscle blood flow during exercise in normal man studied by the 133Xenon clearance method, Cardiovasc. Res. 5:245-254, 1971. 4. Cohen, L. S., Rosenthal, J. E., Horner, D. W., Jr., Atkins, J. M., Matthews, O. A., and Sarnoff, S. J.: Plasma levels of lidocaine after intramuscular administration, Am. J. Cardio!. 29:520-523, 1972. 5. Conn, H. L., Jr.: Equilibrium distribution of radio xenon in tissue: Xenon-hemoglobin associ-
9.
10.
11.
12.
13.
47
ation curve, J. App!. Physio!. 16: 1065- 1070, 1961. Corman, L. A., Flicklinger, F. W., Sokoloff, J., and Nodine, J. H.: Radioactive Xenon tissue clearance: Standardization for measurement of peripheral blood flow, J. Nuc!. Med. 11:233238, 1970. Goldstein, A.: Biostatistics, an introductory text. New York, 1964, The Macmillan Company, p. 136. Holzman, G. B., Wagner, H. N., Iio, M., Rabinowitz, D., and Zierler, K. L.: Measurement of muscle blood flow in the human forearm with radioactive krypton and xenon, Circulation 30:27-34, 1964. Lassen, N. A., Lindbjerg, J., and Munck, 0.: Measurement of blood flow through skeletal muscle by intramuscular injection of Xenon-133, Lancet 1:686-689, 1964. Loken, M. K., and Westgate, H. D.: Using Xenon-133 and a scintillation camera to evaluate pulmonary function, J. Nucl. Med. 9:45-50, 1967. Schwartz, M. L., Meyer, M. B., Covino, B. G., Narang, R. M., Sethi, Y., Schwartz, A. J., and Kamp, P.: Antiarrhythmic effectiveness of intramuscular lidocaine; influence of different injection sites, J. Clin. Pharmacol. 14:77-83, 1974. Smith, G. N., Mollison, D., Griffiths, B., and Mollison, P. L.: Uptake ofigG after intramuscular and subcutaneous injection, Lancet 1: 12081212, 1972. Steel, R. G. D., and Torrie, J. H.: Principles and procedures of statistics, New York, 1960, McGraw-Hill Book Co., p. 107.
