Glycoalkaloid Levels in Potato Products - January-March 1998
January - March 1998
- To determine glycoalkaloid levels in potato products, including skin-on crisps and baked potato peels, sold in the ACT.
- To assess the concentration of glycoalkaloids relative to levels generally regarded as safe.
The cultivated potato (Solanum tuberosum) is consumed daily by many people, forming a staple part of our diet. It is increasingly incorporated in prepared or manufactured foods. The success of the potato as a starch staple has recently been enhanced by breeding pest and disease resistance. However, early domestication was probably more concerned with reducing or eliminating the levels of glycoalkaloids, which are natural bitter-tasting steroidal toxicants (Ref_3). It is assumed the plant synthesizes these materials as a form of protection against viral or bacterial attack, or in response to other biological and physical stresses. Excessive levels of glycoalkaloids may be acutely toxic (causing vomiting, abdominal pain, diarrhoea and general malaise) or, can have teratogenic effects (Ref_4,5).
In potatoes, solanine and chaconine are the two major glycoalkaloids found, accounting for approximately 95% of total glycoalkaloid content. Concentrations are normally less than the widely accepted safety limit of 200 mg/kg fresh weight, however surveys have shown values in excess of this limit for 2 - 9 % of samples. As glycoalkaloids are mainly concentrated in the outer 1.5 mm layer of commercial tubers, peeling removes most of these compounds. Levels in excess of the 200 mg/kg have been reported for potato skin preparations and potato crisps made from unpeeled potatoes. Cooking does not inactivate potato glycoalkaloids (Ref_3,4).
Generally, naturally occurring levels of glycoalkaloids in commercially available tubers are not believed to represent a human health hazard. However, pre- and post-harvest factors can significantly increase levels in the tuber. For example, a 1986 crop of the variety of Magnum Bonum in Sweden was withdrawn due to high level glycoalkaloids caused by a cool and wet growing season (Ref_3). Other factors affecting glycoalkaloid concentrations include potato size (small potatoes have a large surface area), physical damage (cutting and bruising), exposure to light (which can independently stimulate chlorophyll synthesis leading to greening), or microbial and/or herbivore attack (Ref_3, 4, 5). Such factors combined with skin-on or peel based product use, could pose potential health risks (Ref_3).
During 1996/7, a preliminary survey was conducted to check glycoalkaloid levels in raw fresh potatoes sold in Canberra. All of the samples tested showed glycoalkaloid levels below 200 mg/kg fresh weight. However, concentrations found indicated that skin-on or peeled based products should also be monitored and checked for glycoalkaloid levels.
For this survey, a total of 60 samples were collected and tested. 47 of these were potato foods prepared for immediate consumption and the remaining 13 were prepackaged, frozen or dried products purchased in supermarkets.
The following graph shows the relative proportion of sample categories collected for testing.
The greatest proportion of samples tested were hot chips (26) due to public consumption patterns. These sample types are machine peeled potatoes that are sliced and then cooked. As these potato products are peeled expectations were that glycoalkaloid levels would be low.
The next largest sample category tested consisted of potato wedges or skin-on crisps. These products have increased in popularity with consumers over the last few years. They are generally prepared with skin-on and, consequently, were expected to have higher glycoalkaloid levels than hot chips.
Other common potato products tested were hot potatoes and mashed potatoes. Hot potatoes are usually oven or foil baked with their skin-on. The mashed potatoes samples included 2 from a fast food chain and 3 samples of instant mashed potato. Generally these are peeled potatoes and levels of glycoalkaloids were expected to be low.
The remaining sample types consisted of a small number of hash browns, potato scallops, Pommes Noisettes, crunchy potato cubes, and potato gems.
The level of glycoalkaloids found in these foods are summarized in the following graph.
Significantly 31 of the 60 samples analysed (52%) contained no detectable levels of glycoalkaloids. The remaining samples had glycoalkaloid levels below 50 mg/kg with one exception of a potato wedge sample containing 62.5 mg/kg.
A comparison of results between hot chips and wedges, identified a significant difference between the levels of glycoalkaloids detected. For hot chips, the mean concentration of glycoalkaloids was 6.5 mg/kg, ranging from 0 to 27.3 mg/kg, while the wedges had a mean of 25.7 mg/kg with a range of 0 to 62.5 mg/kg.
It was interesting to note that no samples were identified with elevated glycoalkaloid levels. However, testing predictably demonstrated a higher mean concentration in potato wedges than for hot chips due to the wedges being skin-on products.
The levels found in this survey were not of the magnitude indicated by Smith et al (Ref_3). Generally all the different potato foods tested, including skin-on products, were found to have acceptable levels of glycoalkaloids and are not believed to pose a health concern.
No safety concerns were identified in this survey regarding glycoalkaloid levels in potato products, sold in the ACT market. Importantly, this included skin-on crisps and baked potato products.
Refer to recommendations document.
Australia and New Zealand Food Authority, Food Standards Code, incorporating amendments up to and including Amendment 38, April 1998.
Food Act 1992 (ACT), reprinted as at 31 January 1996.
DB Smith, JG Roddick & JL Jones, Potato glycoalkaloids: Some unanswered questions, in Trends in Food Science & Technology, Vol 7, pp126-131, April 1996.
J Hopkins, The glycoalkaloids: Naturally of interest (but a hot potato), in Food Chemistry Toxicology, Vol 33, No 4, pp 323-339, 1995.
J deVries (Editor), Food Safety and Toxicology, pp153-155, CRC Press, 1997.
Sampling Officers: Louisa Bartolome, Chris Wixon and Helen Kivela
Sample analysis: Fiona Wojtas and Andrew Rigg
Report: Simon Christen