• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Food

  • Print
  • Share
  • E-mail

Letter Concerning Arsenic in Pear Juice Products

Date: April 8, 2008

From: Chemical Hazard Assessment Team, Office of Food Safety (HFS-301)

Attn: Regional Milk Specialists

Subject: Arsenic in Pear Juice Products

To: Henry Kim, Ph.D. Office of Food Safety (HFS-300)

As requested, the potential health hazard of the presence of arsenic (As) in pear juice products was considered. 

Estimates of fruit juice consumption were derived from results of the two most recent national food consumption surveys:  the 1994-98 Continuing Survey of Food Intakes by Individuals (CSFII) and the 2003-04 National Health and Nutrition Examination Survey (NHANES).  Although the NHANES is more recent, consumption estimates from the CSFII were also considered because of the larger sample size.  Both surveys included two days of consumption records for most survey participants. 

Although the issue at hand is dietary exposure to arsenic from pear juice/concentrate, information on pear juice consumption was limited in both the CSFII and the NHANES.  Only two food codes were reported for pear juice (infant/toddler pear juice and pear nectar) and the number of consumers of either product was too few to allow the calculation of upper percentiles of consumption.  As an alternate approach, consumption estimates taking into account consumption of all fruit juices were used as a surrogate for consumption of pear juice.

Consumption estimates were based on average consumption over two days for individuals who consumed juice (eaters only) and were calculated for two population groups:  Males and females (MF) from birth to 2 years of age and MF 2 years of age and older.  The consumption estimates are summarized in Table 1. 

Since consumption estimates from both surveys are similar (Table 1), results from the more recent 2003-04 NHANES were used to calculate levels of concern (LOC).  Based on the estimated consumption of all fruit juices (2003-04 NHANES database) and the established safe/tolerable levels of exposure to inorganic arsenic for short-term exposure (the provisional minimal risk level, MRL, of 0.005 mg/kg body weight/day) by CDC/ATSDR and for chronic exposure (Reference dose, RfD), of 3 x 10-4 mg/kg body weight/day) by EPA, the levels of concern of inorganic arsenic in fruit juices are presented in Table 2. 

As shown in Table 2, for short-term exposure, arsenic concentrations in fruit juices at 262 and 115µg/kg are the LOCs for the average and the 90th percentile consumers, respectively, for MF birth to 2 years of age.  For chronic exposure, using the consumption surrogate for MF 2 years of age and older, inorganic arsenic levels below 50 and 23 µg/kg in fruit juices would not pose a health concern for the mean and the 90th percentile juice consumers, respectively.  These levels are in good agreement with the EPA drinking water and FDA bottled water standards of inorganic arsenic of 10 µg/L.  Even though arsenic concentrations are 5 and 2.3 times higher, respectively, for average and 90th percentile consumers, juice consumption levels (221 and 449 ml/day, respectively) are correspondingly lower than drinking water intake (1-2L/day). 

The Total Diet Study (TDS) is an ongoing program that involves the collection and analysis of about 280 different foods four times each year.  For each sample collection (or market basket), three samples of each TDS food are collected in three different cities within the region; the three samples are composited prior to analysis.  Each TDS food composite is analyzed for about 300 different analytes, including toxic elements (total arsenic, cadmium, lead, and total mercury).  Among the foods sampled in the TDS are various fruits juices, including two baby food juices.  Electronic records of TDS analytical results are available from the third market basket of 1991 (MB 91-3) through the fourth market basket of 2005 (MB 05-4).  Most fruit juices were sampled in 51 market baskets between 1991 and 2005; three juices were added to the TDS in 2003 and were sampled 12 times between 2003 and 2005. 

Results for total arsenic levels in TDS samples of fruit juices are summarized in Table 3.  The Level of Detection (LOD) for arsenic in fruit juices is 0.008 mg/kg.  Overall, only 12% of fruit juice samples had detectable levels of arsenic.  Mean arsenic levels, assuming a value of zero for samples below the LOD, ranged from 0 to 13 µg/kg.  Only 4 juice samples had total arsenic levels (24, 25, 29 and 44 µg/kg) above the Level of Concern (LOC) of 23 µg/kg for chronic exposure.  This represents 0.9% (=4/443) of the total samples or 7% (=4/55) of the samples with detected arsenic levels. 

In addition to the TDS samples collected in the TDS, FDA also collected samples of pear juice products in March 2008.  Total arsenic levels found in these pear juice samples are shown in Table 4.  There are a total of 9 samples of finished juice (as consumed) and 12 samples of juice concentrate/blend.  In the finished juice samples, total arsenic levels range from non-detected to 66 µg/kg.  In the pear juice concentrate/blend samples, total arsenic levels when calculated on a finished-juice basis range from 6.5 to 87.9 µg/kg.  Since chemical forms of arsenic in these pear juice products are not available, we assumed that all arsenic is inorganic arsenic.  Of the 9 finished pear juice product samples, 6 (67%) had total arsenic levels above the chronic LOC.  Of the 12 samples of pear concentrate/blend, 9 (75%) had values as calculated on a finished-juice basis that exceeded the chronic LOC.  None of samples had levels that exceeded the short-term LOC. 

In conclusion, the chronic consumption of pear juice products containing over 23 µg/L (ppb) inorganic arsenic would represent a potential health risk. 

/ S /
P. Michael Bolger, Ph.D., D.A.B.T.

/ S /
Katie Egan, B.S.

/ S /
Shirley S.-H. Tao, Ph.D.

Table 1.  Fruit Juice Consumption in the US

Survey/
population

 

# of eaters (unweighted)

 

# eaters

 

2-day average consumption –
eaters only
(g/kg bodyweight/day)

Mean

90th %ile

1994-98 CSFII:

   M/F birth to 2 yrs

1,057

47%

17.6

34.5

   M/F 2+ yrs

3,935

13%

5.8

13.1

2003-04 NHANES:

   M/F birth to 2 yrs

380

56%

19.1

43.4

   M/F 2+ yrs

1,669

15%

6.0

13.0

Table 2.  Levels of Concern of Arsenic in Fruit Juice

 

2-day average
Eaters only (g/kg bw/day)

Level of Concern (LOC)*
As in Fruit Juice, µg/kg (ppb)

Population group

 

mean

90th %ile

mean

90th %ile

MF birth to 2 years

19.1

43.4

262a

115a

MF 2+ years

6.0

13.0

50b

23b

*Level of Concern (LOC) is calculated by dividing the selected safe exposure level of arsenic by the estimated juice consumption (NHANES database).  aFor MF from birth to 2 years of age, MRL of 0.005 mg/kg bw/d for short-term exposure is used.  bFor MF 2 years of age and older, RfD of 3 x 10-4 mg/kg bw/day for chronic exposure is used. 

Table 3:  Arsenic Concentration in Fruit Juices: TDS results from 1991 through 2005

Food Description

total n

#
detects

# of non-
detects

% non-
detects

As concentration (µg/kg)

Mean
(ND=O)

Min

Max

Baby food, apple juice

51

8

43

84%

3

0

24

Baby food, orange juice

50

0

50

100%

0

0

0

Orange juice, from frozen concentrate, reconstituted

51

2

49

96%

1

0

19

Apple juice, bottled

51

14

37

73%

4

0

40

Grapefruit juice, from frozen concentrate, reconstituted

51

1

50

98%

0.2

0

11

Prune juice, bottled

51

3

48

94%

1

0

14

Pineapple juice, from frozen concentrate, reconstituted

51

1

50

98%

0

0

25

Grape juice, from frozen concentrate, reconstituted

51

16

35

69%

4

0

23

Fruit juice blend (100%
juice), canned/bottled

12

9

3

25%

13

0

44

Cranberry juice cocktail, canned bottled

12

1

11

92%

1

0

9

Orange juice, bottled/carton

12

0

12

100%

0

0

0

Table 4:  Arsenic Levels in Pear Juice Products Sampled by FDA

Sample #

Product

Arsenic Concentration
µg/kg (ppb)

In juice
Concentrate

In finished juice
(as consumed)

Pear Juice

421576

Pear juice

 

Not detected

421577

Pear juice

 

Trace

434571

Pear juice

 

13

438968a

Pear juice from
concentrate, 4 oz.

 

62.1-63.9

438971a

Pear juice from concentrate, 4 oz.

 

66.0

438972a

Pear juice from concentrate, 32 oz.

 

33.6

453798

Pear juice from concentrate

 

52

453799

Pear juice from concentrate

 

34

443800

Pear juice from concentrate

 

64

Pear Concentrate/Blend

453801

Pear juice concentrate

53

7

455152b

Pear juice conc. Blend of 3 lots

360

61.6

455153

Pear juice concentrate

420

71.9

455154

Pear juice concentrate

209

35.5

455155c

Pear juice conc. Blend of 2 lots

299

50.9

455156

Pear juice concentrate

523

87.9

455157

Pear juice concentrate

65.6

11.2

461043

Pear juice concentrate

430

73.6

461044

Pear juice concentrate

360

61.1

461837

Pear juice concentrate

123

21.1

461915

Pear juice concentrate

37.6

6.5

462012

Pear juice concentrate

69

8.6

All samples were collected in March, 2008. aProduced with concentrate sample #461043.  bBlend of 3 lots (including samples #s 455153/4).  cBlend of 2 lots (sample #s 455156/7).