The acute lethal dose 50 (LD50) of caffeine in albino rats

Accepted Manuscript The acute lethal dose 50 (LD50) of caffeine in albino rats Richard H. Adamson PII:

S0273-2300(16)30194-5

DOI:

10.1016/j.yrtph.2016.07.011

Reference:

YRTPH 3633

To appear in:

Regulatory Toxicology and Pharmacology

Received Date: 23 December 2015 Revised Date:

15 July 2016

Accepted Date: 18 July 2016

Please cite this article as: Adamson, R.H., The acute lethal dose 50 (LD50) of caffeine in albino rats, Regulatory Toxicology and Pharmacology (2016), doi: 10.1016/j.yrtph.2016.07.011. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

M AN U

Commentary

SC

RI PT

ACCEPTED MANUSCRIPT

The Acute Lethal Dose 50 (LD50) of Caffeine in Albino Rats

AC C

EP

TE D

Richard H. Adamson TPN Associates LLC, 13625 Esworthy Road., Germantown, MD 20874 and 7 Haynes Street, Walpole, MA 02081, United States Email address:[email protected]

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

The Acute Lethal Dose 50 (LD 50) of Caffeine in Albino Rats

1

Abstract: An acute LD50 is a statistically derived amount of a substance that can be expected to cause death in 50% of the animals when given by a specified route as a single dose and the

TE D

animals observed for a specified time period. Although conducting routine acute toxicity testing in rodents has been criticized, it can serve useful functions and also have practical implications. Material safety data sheets (MSDS) will reflect the acute toxicity of a substance and may require

EP

workers to wear protective gear, if appropriate, based on the LD50. There is no information in the scientific published literature which calculates a mean

AC C

LD50 and standard deviation for caffeine administered orally to rats, using studies performed under good laboratory practice (GLP) or equivalent. This report does that and should be useful to manufacturers, packagers, transporters and regulators of this material. Using data from studies that are reproducible and reliable, the most accurate estimate of the acute LD50 of caffeine administered orally in male albino rats is hereby reported to be 367/mg/kg.

ACCEPTED MANUSCRIPT

Although safety and efficacy of any medicine, food component, food additive, dietary

3

supplement, chemical, or substance are important, safety is of paramount importance. But, as

4

Philippus Aureolus Theophrastus Bombastus von Hohenheim, known as Paracelsus and often

5

referred to as the father of toxicology, stated, “Poison is in everything and nothing is without

6

poison. The dosage makes it a poison or a remedy.” Today, one would say everything is toxic at

7

some dose level.

10

SC

9

The toxicity of a substance, prior to use in humans, can be determined by use of in vitro tests, in silico studies and studies in animals, especially in rodents and also in other species.

M AN U

8

RI PT

2

The acute lethal dose 50 (LD50) toxicity test is performed in rodents as part of the safety assessment of many substances. The LD50 is a statistically derived amount of a substance that

12

can be expected to cause death in 50% of the animals when given by a specified route as a single

13

dose and the animals observed for a specified period of time (Chan and Hayes, 1989). Although

14

conducting routine acute toxicity testing has been criticized, it can serve useful functions.

15

Differences in the LD50 in different species might yield clues to differences in absorption,

16

metabolism and mechanism of toxicity of a substance. The LD50 in a given species is not

17

constant due to differences in factors (Boyd, 1968; Branton, 1993; Peters, 1967)such as fasted or

18

nonfasted state, type of diet/nutritional state, route and rate of administration, the nature of the

19

vehicle in which the substance was dissolved, the volume of the vehicle in which the substance

20

was dissolved, the age, sex, weight and strain of the animals, the purity of the substance, the

21

number of animals used, the nature and cause of death and the length of time the animals are

22

observed. Thus, the LD50 usually spans a range of values and the most accurate and reproducible

23

value depends on defining the factors in the study, the statistical method used and conducting

AC C

EP

TE D

11

ACCEPTED MANUSCRIPT

24

and reporting the study so that the circumstances in which the study was conducted can be

25

replicated. The LD50 of a substance also has practical implications. Material safety data sheets

27

(MSDS) will reflect the acute toxicity of a substance and require workers to wear protective gear,

28

based on the LD50. OSHA’s acute toxicity hazard categories have strict cutoffs (5, 50, 300, and

29

2000/mg/kg) which means the difference in the calculated LD50 can greatly affect protection

30

required. (U.S. Department of Labor Occupational Safety and Health Administration

31

Regulations.) Likewise, the acute LD50 of a substance also affects labeling and packaging of

32

substances in transport. For example, in the United Nations (UN) recommendations on the

33

Transport of Dangerous Goods, the criteria for assignment of a packing group to a Class 6.1

34

substance are based on LD50 values of 300 mg/kg or less. (United Nations, 2011; and United

35

Nations, 2010)

SC

M AN U

TE D

36

RI PT

26

The accuracy of the LD50 of caffeine reported by the UN is questionable. The current assumption is that it is below 300 mg/kg, but a full evaluation of all the available data does not

38

seem to have been performed. The United Nations report lists caffeine in Group 6.1 and this has

39

been adopted in U.S. Code of Federal Regulations as a liquid or solid with an LD50 for acute

40

oral toxicity of not more than 300 mg/kg (United Nations, 2010 and U.S.49 Code of Federal

41

Register§173.132 Class 6, Division 6.1). An inaccurate classification of caffeine puts an

42

unnecessary burden on the pharmaceutical, food and beverage industries from an operational

43

standpoint. The LD50 value used for assignment of a classification must take into consideration

44

the factors and corresponding reporting in various studies determining the LD50 and be based on

45

the best performed and reported studies, especially if conducted under GLP or GLP equivalent

46

standards rather than on a specific study or a summary and/or average of all the literature values.

AC C

EP

37

ACCEPTED MANUSCRIPT

47

The author reviewed 15 reports, three are unpublished reviews ( Branton,1993; Cohen et al.,2006; Young and Watts, 2015) and eleven are original reports in the published scientific

49

literature ( BASF,1985; Bruce,1987; Boyd,1959; Boyd et al.,1968; Dashiell and Kennedy,1984;

50

NTP,1982a; NTP 1982b; Hasegawa,et al. 1989; Palm et al.,1978; Schlede et al.,1992; and Scott

51

and Chen, 1944) and a United Nations Environmental Programme ( UNEP ) publication on

52

caffeine ( OECD, 2002). Some experimental studies on caffeine’s acute toxicity were performed

53

prior to today’s standards for performing and reporting studies for regulatory purposes.

SC

RI PT

48

The three review articles, not published, summarized the literature on the LD50 of

55

caffeine up to 1993, 2006 and 2015 ( Branton,1993; Cohen et al.,2006; Young and Watts, 2015).

56

No experimental studies have been published in the literature since 1992 ( Schlede et al.,1992).

57

The review by Branton (1993) concluded that “based on the weight of evidence we conclude

58

that the oral LD50 of caffeine in rats lies in the range of 200-400 mg/k of body weight.” The

59

report by Cohen et al. (2006) stated that “the weight of evidence from the various studies

60

supports the conclusion that the acute oral LD50 for male rats is greater than 300 mg/kg. The

61

limited useful information available and the lack of confirmatory studies performed under the

62

same study conditions make it impossible to draw a firm conclusion for the female rat LD50.”

63

Young and Watt (2015) report in their summary stated that “several reported rat acute oral LD50

64

values have been identified and these ranged from 50-1050 mg/kg bw. Most of the identified oral

65

LD50 values are between 200-420 mg/kg bw, and this range includes the evidently highest

66

quality and most reliable studies (Klimisch 1 and 2)”. None of the three reviews using the best

67

reported data calculated a mean LD50 and standard deviation for caffeine administered orally to

68

rats. This information is needed by regulatory agencies to make a best informed decision on

69

packaging, labeling and worker safety.

AC C

EP

TE D

M AN U

54

ACCEPTED MANUSCRIPT

Published studies reviewed are referenced above and OECD (published by UNEP) also

71

did a literature summary on animal data, the human health effects, exposure and environmental

72

effects of caffeine ( OECD, 2002 ). This 375 page report did not report any calculations on the

73

LD50 of oral caffeine in rats but just reported the LD50 values (range) from the NTP 1982b

74

study and the BASF 1985 study. Table I summarizes the studies on the acute LD50 of orally

75

administered caffeine in albino rats. Young and Watts (2015 ) assessed the reliability of each of

76

these studies using the Klimisch criteria ( Klimisch et.al.,1997). The current author also provided

77

an independent assessment based on the Klimisch criteria, similar to that of Young and Watts,

78

that also yielded studies for packing group labeling for transportation purposes (GLP or

79

equivalent).

SC

M AN U

80

RI PT

70

The six studies with a Klimisch criteria ( Quality ) of 1 or 2 that were also judged to be useful for labeling for transportation purposes and could be reproduced from their publication are

82

briefly summarized in Table 1. These six studies stated the strain of rat used, the sex of animal

83

used, the vehicles used for caffeine administration, the fasting time of the animal (no more than

84

18 hours), the time of observation period, and the statistical method used for LD50 calculation –

85

data that allowed for replication of the study. The six studies ( BASF 1985; Bruce 1987; NTP

86

1982a; NTP 1982b; Palm et al.,1978) also provided a LD50 value for male rats. The values of

87

these six reports, a calculated mean of the six and a standard deviation are : 400, 300, 383, 355,

88

349, 412 mg/kg ; mean ± SD =367 ± 37 mg/kg. Limited useful information is available for the

89

female rat LD50 . The value is from only four studies potentially useful for labeling (

90

BASF,1985;NTP,1982a;NTP, 1982b, Palm et al.,1978 )-two of which are best estimates-

91

(NTP,1982a; NTP,1982b ). These do not allow for any firm conclusion to be made for female

92

rats but an estimate ( average ) from these four studies would be greater than 300 mg/kg.

AC C

EP

TE D

81

ACCEPTED MANUSCRIPT

Thus, using data from studies that are reproducible and reliable, the most accurate estimate of the

94

acute LD50 of caffeine administered orally in male albino rats is 367/mg/kg.

95

Summary

96

RI PT

93

The LD50 of a compound in a given species provides useful pharmacologic and

toxicologic information and practical information relative to worker protection and labeling

98

classification for transportation. However, the LD50 information should be based on the best

99

conducted and reported studies, especially those performed under GLP or equivalent conditions.

SC

97

Thus, the LD50 data for caffeine used by the United Nations should be reviewed, and a more

101

accurate LD50 used, based on the best performed and reported studies. A review of the studies

102

performed on the LD50 of caffeine supports that the value from the studies of NTP (

103

1992a;1992b), BASF ( 1985) , Bruce ( 1987) and Palmet al. ( 1978) are of appropriate quality

104

(Klimisch scores of either “1“ or “2”) and also judged useful for transportation classification

105

labeling purposes. The most accurate estimate of the acute LD50 in rats is 367 mg/kg. Other

106

studies are of lesser quality and are therefore not pertinent, and should not be the basis for

107

labeling.

TE D

EP AC C

108

M AN U

100

ACCEPTED MANUSCRIPT

109

110

References

RI PT

111 112

BASF (1985). Test for the acute oral toxicity of the rat. Unpublished data (cited in OECD 2003).

115 116

Boyd, E.M. (1959). The acute oral toxicity of caffeine. Toxicology and Applied Pharmacology 1, 250-257.

117 118 119 120

Boyd, E.M. (1968). Predictive drug toxicity:assessment of drug safety before human use. Canad. Med. Ass. J. 98, 278-293.Boyd, E.M., Liu, S.J., Singh, J. (1968). The toxicity of aspirin, phenacetin, and caffeine following rectal administration. Clinical Toxicology 1(4), 425-430.

121 122 123 124 125

Branton, Paul. G., (1993). Review of the available data on LD50 of caffeine given orally to rats. Unpublished, personal communications.

126 127 128 129 130 131

Chan, P.K. and Hayes, A.W. (1989). Principles and Methods of Acute Toxicity and Eye Irritancy, Principles and Methods of Toxicology. Edited by A. W. Hayes, Raven Press, Ltd. New York.

132 133

Dashiell, O.L. and Kennedy, Jr. G.L. (1984). The effects of fasting on the acute oral toxicity of nine chemicals in the rat. J. Applied Toxicology 4, 320-325.

134 135 136

Hasegawa, R., Nakaji, N., Kurokawa, Y. and Tobe, M. (1989). Acute toxicity tests on 113 environmental chemicals. Science Reports of the Research Institutes Tohoku University 36, 81-4, 10-16.

137 138 139

Klimisch, H.J., Andreae, M. and Tillmann, U. (1997). A systematic approach for evaluating the quality of experimental toxicological and ecotoxicological data. Regulatory Toxicology and Pharmacology 25, 1-5.

140 141 142

NTP (1982a). National Toxicology Program, National Center for Toxicological Research (Jefferson, AK September 1982). Acute Toxicity Report on Caffeine in Fischer 344 Rats, NCTR Experiment No.365.

M AN U

SC

113 114

TE D

Bruce, R.D. (1987). A confirmatory study of the up-and-down method for acute oral toxicity testing. Fundamental and Applied Toxicology 8, 97-100.

AC C

EP

Cohen, S.M., Galbraith, W.M. and Munro, J.C., (2006); Report on caffeine acute oral LD50 in albino rats. Unpublished, personal communication.

ACCEPTED MANUSCRIPT

NTP (1982b.) National Toxicology Program, National Center for Toxicological Research (Jefferson, AK September 1982) Acute Toxicity Report on Caffeine in Fischer 344 Rats, NCTR Experiment No. 374.

146

OECD, SIDS, UNEP Publications 2002 Caffeine (CAS :58-08-2) 1-376.

147 148 149

Palm, P.E., Arnold, E.P., Rachwall, P.C., Leyczok, J.C., Teague, K.W. and Kensler, C.J. (1978). Evaluation of the teratogenic potential of fresh-brewed coffee and caffeine in the rat. Toxicology and Applied Pharmacology, 44, 1-16.

150 151 152 153

Peters, J.M. (1967). Factors affecting caffeine toxicity. J.Clin. Pharm. May-June, 121141.Schlede, E., Mischke, U., Roll, R. and Kayser, D. (1992). A national validation study of the acute-toxic-class method – an alternative to the LD50 test. Archives of Toxicology, 66, 455-470.

154 155

Scott, C.C. and Chen, K.K. (1944). Comparison of the action of 1-ethyl theobromine and caffeine in animals and man. (1944). J. Pharmacology and Experimental Therapeutics. 89-97.

156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176

United Nations (2010). Classification, 2, New York and Geneva: United Nations, retrieved from https://www.unece.org/fileadmin/DAM/trans/danger/publi/adr/adr2011/English/Part2.pdf p.197, Section 2.2.61.1.7

178 179 180

SC

M AN U

TE D

United Nations. (2011). Recommendations on the Transport of Dangerous Goods, 1 New York and Geneva: United Nations, retrieved from http://www.unece.org/fileadmin/DAM/trans/danger/publi/unrec/rev17/English/Rev17 Volume1.pdf , p.376 and 342. U.S.49 Code of Federal Register §173.132 Class 6, Division 6.1

EP

U.S. Department of Labor, Occupational Safety and Health Administration Regulations (Standards) 29 CFR part 1910, subpart z, standard 1910.1200 App. A also found at osha.gov/pls/oshaweb/ Young, R. and Watts, P. (2015). Acute oral LD50 values of caffeine in rats. Unpublished, personal communication.

AC C

177

RI PT

143 144 145

ACCEPTED MANUSCRIPT

181 182

Conflict of Interest Information

184 185

RI PT

183

The funding for the literature review of this article was provided by the American Beverage Association.

188

Richard H. Adamson is a consultant for the American Beverage Association.

SC

186 187

189

M AN U

190

AC C

EP

TE D

191

ACCEPTED MANUSCRIPT

Charles River (Cri-CD)

Equivalent to OECD Guideline 401 conducted by FDA/NCTR for NTP. Rats fasted 18 hours, observed for 14 days Equivalent to OECD Guideline 401 Conducted by FDA/NCTR for NTP. Rats fasted 18 hours, observed for 14 days Comparable to EPA guidelines but fasted for 24 hours instead of 16 hours. Caffeine suspended in corn oil

SAS probit

Equivalent to GLP

Quality

1

SC

GLP or Equivalent

M AN U

M/F

Calculated Method

SAS probit

TE D

F344 6/sex/group

M/F

Brief description of study

Probit analysis

EP

F344 6 /sex/group

Sex

AC C

Strain and Number of Rats

RI PT

Table I. The Acute Oral LD50 of Caffeine in Albino Rats

Equivalent to GLP

No

Useful Reference in Labeling Yes NTP (1982a)

LD50 mg/kg 400 best estimate

1

Yes

NTP (1982b)

300 (M) 420 (F) estimate

3

No

Dashiell and Kennedy 1984

279 (M) fasted 483 (M) not fasted

Table 1 (continued)

M/F

Sprague-Dawley 40 rats 10/group Sprague-Dawley 1 rat/dose level 9 rats

M

M

Standard of acute toxicity method (“BASF”) similar to OECD Guideline 401. Rats observed for 14 days. Rats fasted for 16 hours prior to caffeine dosage observed for 15 days Fasted 18-20 hours Observed 15 days

SAS probit

Equivalent to GLP

Up and down method vs. classical method

Cornfield and Mantet

Quality

SC

Sprague-Dawley

GLP or Equivalent

M AN U

M/F

Calculation Method

1

Reference Useful in Labeling Yes BASF 1985

LD50 mg/kg 383 (M) 281-383 (F)

Unknown

2

Yes

Palm et al. 1978

355 (M) 247 (F)

SAS Probit

Unknown

2

Yes

Bruce 1987

349 (M)

SAS Probit

Unknown

2

Yes

Bruce 1987

412 (M)

TE D

Wistar 5/sex/group

Brief description of study

EP

Sex

AC C

Strain and Number of Rats

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Table I (continued)

Young Wister rats M 6-10/group

Sprague-Dawley or Wistar 3 rats/group

M/F

Sprague-Dawley 10/group

M

Footnote

4

No

Boyd 1959

192± 18

4

No

Boyd et al. 1968

300 anesth. 288± 6 Non-anesth.

No

4

No

Schlede et al.

200-2000

No

4

No

Hasegawa et al.

700

Method of Bliss

No

Not given

No

Fasted for 16 hours LD50 of both nonanaesthetized and anaesthetized rats West German validation study of acute-toxic- class as alternative to conventional LD50. Doses of 25, 200 and 2000 mg/kg not stated if fasted Fasted, administered in gum Arabic by gavage

Not given

No

Not given

RI PT

F

The median lethal dose was determined in starved albino rats 8 rats/ dose 3-6 months age

Quality

4

SC

Albino rats

GLP or Equivalent

M AN U

M/F

LD50 mg/kg

Calculation Method

233 ± 14

TE D

Albino rats 9/dose

Reference Useful in Labeling No Scott and Chen

Brief description of study

EP

Sex

Not given

AC C

Strain and Number of Rats

Table after Richard Young and Pete Watts (2015) Acute oral LD50 values of caffeine in rats, June 2015 The criteria from Klimisch et al. (1997) as used by Young and Watts (2015) was 1 (reliable without restrictions; e.g. guideline studies); 2 (reliable with restrictions; e.g. well-documented publications); 3 (not reliable; e.g. methodological deficiencies); 4 (not assignable; e.g. not sufficient experimental details, secondary literature only)

ACCEPTED MANUSCRIPT

Highlights

The lethal dose 50 (LD50) of a substance to rodents is useful and may have practical implications.

•

The LD50 used for regulatory purposes should be based on the best performed studies.

•

The most accurate estimate of the LD50 of caffeine orally to rats is 367mg/kg.

AC C

EP

TE D

M AN U

SC

RI PT

•