Caballero B. (ed.) Encyclopaedia of Food Science, Food Technology and Nutrition. Ten-Volume Set

Подождите немного. Документ загружается.

the applications of the different microscopes in this

family extend to provide information on physical as

well as structural properties, including information

on topography, hardness, capacitance, stiffness/adhe-

sion, and microcalorimetry. For some sample types

the resolutions quoted are to < 1 nm and applications

have been demonstrated in biology, pharmaceutical,

and metallurgical industries. The techniques have

potential for applications for foods and have already

been used to demonstrate the structures of proteins in

some food gels and emulsions.

See also: Analysis of Food; Carbohydrates:

Determination; Microscopy: Scanning Electron

Microscopy; Protein: Determination and

Characterization; Sensory Evaluation: Sensory

Characteristics of Human Foods

Further Reading

Aguilera JM (1990) Microstructural Principles of Food

Processing and Engineering 1. Food Science. Barking:

Elsevier.

Chescoe D and Goodhew P (1982) The Operation of the

Transmission Electron Microscope. Royal Microscop-

ical Society, Microscopy Handbooks No. 02. Oxford:

BIOS Scientific Publishers.

Hayat MA (1989) Principles and Techniques of Electron

Microscopy; Biological Applications. London: Macmillan.

Jewell GG and Heathcock JF (1988) Structured fat systems.

In: Blanshard R and Mitchell J (eds) Food Structure: Its

Creation and Evaluation, pp. 279–296. London: Butter-

worth.

Kalab M (1983) Electron microscopy of foods. In: Peleg

M and Bagley EB (eds) Physical Properties of Foods,

pp. 43–104. Westport: AVI Publishing Co.

Polak J and Van Noordan S (1984) An Introduction

to Immunocytochemistry: Current Techniques and

Problems. Royal Microscopical Society, Microscopy

Handbooks No. 11. Oxford: BIOS Scientific Publishers.

Salyaev RK (1968) A method of fixation and embedding of

liquid and fragile materials in agar microcapsules. Pro-

ceedings of the Fourth European Conference of Electron

Microscopy 11: 37–38.

Vaughan JG (ed.) (1979) Food Microscopy. London:

Academic Press.

Image Analysis

A Flook, Bedford, UK

Introduction

0001Image analysis is a computer-based process of

extracting quantitative information from images.

The process begins with the input of an image and

100 nm

fig0004 Figure 4 Freeze-fracture replica of a confectionery emulsion showing the fine emulsified fat droplets (F) and aggregated casein

protein (P). Bar ¼100 nm. Reproduced from Microscopy: Transmission Electron Microscopy, Encyclopaedia of Food Science, Food

Technology and Nutrition, Macrae R, Robinson RK and Sadler MJ (eds), 1993, Academic Press.

3934 MICROSCOPY/Image Analysis

ends with the output of numerical data (Figure 1).

This distinguishes it from image processing where

both input and output are in the form of an image.

0002 Image processing is the means by which the input

image is modified by one or more mathematical algo-

rithms to generate an output image that is enhanced

in some way. For example, edges may be enhanced or

the noise reduced. Image processing is often used to

prepare images prior to analysis.

0003 Image analysis requires specialized computer equip-

ment fitted with an imaging device, such as a television

camera, coupled to a microscope or a macroviewer.

0004 The first commercial image-analyzing computers

were constructed in the early 1960s for grading steel

by measuring nonmetallic inclusions. The value of

using an automated method for the rapid, objective

measurement of images was soon recognized for

many different applications in life and materials

sciences. Image analyzers with greater versatility

have been developed but initially the application of

the technique was limited by the high cost of the

instrumentation. However, the price of computer

hardware has steadily decreased over the years

and now many manufacturers offer relatively

inexpensive systems using a personal computer as

the host processor. With the advent of more afford-

able equipment, applications have proliferated in

many widely differing fields, including food science

and technology.

Terminology and Nomenclature

0005 Image-processing terminology is mainly derived from

the mathematical functions used (e.g., Laplacian and

Gaussian filters are used for image enhancement).

These terms may not always be easy to understand

but are used universally and can, in the main, be used

without ambiguity. This cannot be said for image

analysis. There are no internationally agreed stand-

ards and the terminology used can be specific to the

manufacturer of the equipment. Indeed, in the early

years of image analysis it seemed almost obligatory

for each new analyzer on the market to come with its

own set of terminology. For example, one manufac-

turer uses the term segmentation for the process of

extracting a binary image by gray-level thresholding.

Another calls this process detection and reserves the

term segmentation for the process of separating

touching regions. Also, the binary regions to be meas-

ured can be referred to as regions, objects, features, or

even blobs, depending on the equipment used. This

makes it difficult for newcomers to read research

papers and relate the functions named to those

installed on their own equipment.

0006This situation, although regrettable, is unlikely to

change. The terms used in this article are those most

commonly used but the reader should be aware that

they are not unique.

The Hardware

0007There is a wide range of image analyzers commer-

cially available. They all differ in detail but are

broadly similar in design. A typical system consists

of the following components.

Microscope

0008This is usually fitted with a trinocular head to allow

direct viewing of the specimen and for the mounting

of a television camera. (A macroviewer is an alterna-

tive optical input device for photographs or larger

specimens.)

Television Camera

0009Most image analyzers use standard-format charge-

coupled device (CCD) TV cameras, although an

increasing number of systems are being fitted with

color and high-definition cameras.

Analog-to-Digital Converter (ADC)

0010This is a device that converts the continuous voltage

output signal from the camera into a sequence of

numbers representing the voltage at timed sampling

intervals. The lighter the region of the picture at the

sampling point, the higher the voltage and hence the

larger the numerical value. Most systems are fitted

with an 8-bit converter, allowing 2

¼256 different

gray-levels. Three independent channels are used

for color images – one each for the red, green, and

blue images that constitute a conventional real-color

image.

Data

Image

Image synthesis

Image analysis

Data processing

Image processing

fig0001 Figure 1 Conversion of information between data and image

format.

MICROSCOPY/Image Analysis 3935

Image Store

0011 The sampling interval of the ADC is chosen to give a

discrete number of pixels (an abbreviated term for

picture elements) for each scan line of the television

signal. These are read into an image store. The format

for the image store is a matrix of pixels corresponding

to the CCD format of the camera. The store is usually

large enough to accommodate several digitized

images at once. Systems based on personal computers

use buffers within the random access memory.

Monitor

0012 This displays the images and provides a user interface

for programming. Most image analyzers use a

scripting language to build programs that control

the processing of the images and the analysis of the

data.

Image Processor

0013 Digitized images are very large arrays of numerical

data. Early instruments had specialized image proces-

sors to manipulate the images but modern personal

computers have processors powerful enough to cope

with the demands of image processing without using

special processors.

Binary Image Store

0014 In most applications the image will ultimately be

reduced to a binary form where each pixel of the

original image is converted to either a 1, representing

a pixel within an area occupied by one of the objects

to be measured, or a 0 representing a pixel in the

background. Binary images are normally displayed

either as bright pixels on a dark background or by

the use of two distinct colors representing the 0s

and 1s.

0015 Early instruments had specialized memory stores

for this type of data but these are now obsolete.

The Host Processor

0016 This is often, but not necessarily, a personal com-

puter. On older machines it is used to manage the

data flow within the computer and issue instructions

to the specialized processors. Now all the processing

is usually done using the standard computer proces-

sor. The host processor also supervises peripheral

devices such as printers and the automatic focus and

x,y stage controls on the microscope (where fitted).

Finally, it provides the means to process the data and

provide any statistical, graphical, or other utility pro-

gram that may be required for a particular applica-

tion. The host processor will usually be fitted with

disk drives for the storage of data and programs.

Output Device

0017This will commonly be a printer or plotter that will

enable the results to be converted into hard copy in

graphical or tabular form.

The Principles of Image Analysis

0018Image analysis often requires a greater degree of care

in preparing the sample and setting up the microscope

than is necessary for visual inspection. The human

brain is well adapted to the recognition of objects of

interest in an image and to ignore extraneous infor-

mation. The computer, on the other hand, can only

cope with relatively simple images. For this reason the

image analysis process should be considered to start

from the sample preparation onwards.

0019Measurements can only be made on a binary

image. The ultimate goal of the image prepossessing

is to simplify the image in such a way that will make it

easy to convert it to this form. In the binary image the

bright pixels should correspond to those areas in the

original image that represent the objects or structures

that are to be measured.

0020A typical sequence will include the following steps.

Sample Preparation

0021Small variations in the preparation protocol can de-

termine whether or not a project is amenable to image

analysis. The best images for analysis are those that

show the objects that are to be measured in greatest

contrast with respect to the rest of the image. Color

contrast may be used with those instruments fitted

with a color camera. Intensity contrast is used for

monochrome images.

Image Capture

0022The analysis proceeds by setting up the input device

to deliver the best possible image to the camera.

When a suitable image is obtained, it is read into the

computer memory via the ADC.

Image Processing

0023There are many methods of mathematically process-

ing an image so that the output image is enhanced in a

manner that renders it more suitable for image analy-

sis. For example, the edges of structures may be

sharpened or the image may be given enhanced con-

trast or reduced noise.

0024It is not always necessary to preprocess the image

in this way, but shading correction is almost always

used. Shading correction is the process by which the

computer attempts to compensate for any uneven

illumination of the specimen. This is usually achieved

by storing the pattern of the variation in intensity

3936 MICROSCOPY/Image Analysis

obtained from a blank image field. This image, called

the shading pattern, is used to correct subsequent

images.

0025 Image processing can never increase the informa-

tion content of an image. Any enhancement of

one image attribute is obtained at the expense of

another. For example, edge enhancement algorithms

tend to increase the noise content of an image. Many

algorithms have been developed over the years to

enhance images with the minimum of unwelcome

side-effects. There are, for example, methods of

reducing noise without smoothing out the edges of

objects. However, heavy processing can still intro-

duce unwanted artifacts in the image. For this reason

sample preparation and lighting conditions are best

chosen to reduce the need for image enhancement to a

minimum.

Segmentation

0026 This is the process by which the digitized, gray,

or color image is converted into a binary image

representing the objects to be measured. This is the

most important, and often the most difficult, step in

image analysis. The computer needs a binary image

to make measurements. There are several ways of

defining the regions to be segmented. The simplest

form is by setting a gray-level value below which

(or above, for light objects) a pixel will be counted

as being part of the object. All other pixels will

be counted as belonging to the background. This

form of segmentation is also referred to as threshold-

ing or detection. Another method allows the operator

to draw round or point to a typical region; the com-

puter will sample the intensities of the pixels in the

region and generate a binary image of all regions

containing pixels with matching intensities. There

are also methods that specifically detect the edges of

objects.

0027 Color discrimination provides an alternative

method of segmentation for true color images. There

are two main color models used in image analysis–the

red, green, blue (RGB) model and the hue, luminance,

saturation (HLS) model. Color images are separated

into their three components (RGB or HLS) according

to which color model is to be used for segmentation.

Each component is held in memory as a separate

image. These component images can be viewed separ-

ately as monochrome images but are normally dis-

played combined as a color image (Figure 2).

Segmentation methods, similar to those used on

monochrome images, are applied to each of these

components and the resultant binary images com-

bined. The resultant binary image will represent

those regions in the original image that have a similar

color composition.

Binary Image Processing

0028Often the segmentation process does not produce a

perfect binary image of the objects of interest and

many binary operations have been devised to amend

the defects in the image. For example, there are oper-

ations for filling in holes and others that dilate or

erode the binary image by adding or subtracting

pixels from the perimeter of objects. Erosion and

dilation can often be used to good effect by combin-

ing them to form new operations known as closing

and opening. Closing is the process of dilation

followed by an equal amount of erosion and is useful

for joining up fragmented regions of an object.

Opening is the converse operation and is often used

to clean up ragged edges and remove small or thin

artifacts from the image. Opening and closing oper-

ations leave the overall size of the objects relatively

unchanged compared with erosion or dilation used on

their own.

0029There are other more complex binary operations

which can be used, for example, to separate touching

objects, reduce objects to a skeletal representation,

and erode objects to the point just before they

would otherwise disappear. These tools allow the

image analyst to automate much of the editing of

binary images that would otherwise have to be done

by time-consuming manual methods. They form an

important part of image analysis and can make the

difference between being able or otherwise to per-

form an analysis in a reasonable time. The theory

behind these operations is known as mathematical

morphology.

Editing

0030The binary image can be edited manually by using a

hand-held device, such as a mouse or a light pen,

which is interactive with the computer. By this

means sections or lines can be cut out or drawn in at

will. Although it is often necessary to do a little

manual editing, this should always be regarded as a

method of last resort since it is a slow operation and

introduces a subjective element into the analysis.

Measurement

0031There are two distinct modes of measurement that can

be made in image analysis: field-specific and object-

specific. Before discussing these it is important to dis-

tinguish between the image and measurement frames.

The total area of the specimen scanned by the TV

camera and displayed on the monitor forms the image

frame. The measurement frame is a subframe within

the image frame over which the measurements are

made. The area within the image frame that lies outside

the measurement frame forms the guard region.

MICROSCOPY/Image Analysis 3937

0032 Field-Specific measurements Field-specific meas-

urements are summed over all the objects and parts

of objects lying within the measurement frame. In this

mode, if an object is intersected by the edge of the

measurement frame, only those parts of the object

that project into the frame are measured. Field-spe-

cific measurements are used for stereological calcula-

tions of global parameters such as phase volume or

count per unit volume.

0033 Object-specific measurements Object-specific meas-

urements are made on each individual object in the

measurement field. In this mode, each object is asso-

ciated with a unique object location reference point

that is determined by the counting logic of the par-

ticular image analysis equipment used. The whole of

an object is measured if its reference point is inside

the measurement frame, even if much of it spills over

the edge of the frame. Conversely, the object will not

be measured at all if the reference point falls outside

the frame. Any uniquely definable point may be

chosen as the reference but, for practical reasons,

the location of the last pixel visited within the object

as the scan progresses down the screen is most com-

monly chosen. The measurement frame size is chosen

so that the guard region, which surrounds it, is wide

enough to insure that any object whose location point

is within the measurement frame is wholly within the

image frame.

0034Some primary measurement parameters for objects

are defined below. Most of these measurements can

also be made in field-specific mode:

0035Area: the total number of pixels within an object

0036Perimeter: the length of the boundary between an

object and the background

fig0002 Figure 2 (see color plate 109) Color segmentation of a section from a pea cooked from frozen. (a) Original image; (b) cell walls;

3938 MICROSCOPY/Image Analysis

.0037 Horizontal intercept: the number of intersections

of the TV scan line with an object

0038 Vertical intercept: the equivalent of the horizontal

intercept but in a vertical direction

0039 Feret’s diameters: These are equivalent to caliper

diameters, i.e., the perpendicular distance between

the parallel jaws of a caliper just touching the

object. The magnitude of these measurements is

dependent on orientation and the more advanced

image analyzers are able to measure Feret’s

diameters at a large number of small angular incre-

ments.

0040 Integrated brightness: The gray-level values sum-

med over the region corresponding to the binary

image of the object. With suitable calibration, these

measurements can be converted to optical density

values.

0041 Count: In field-specific mode this gives the total

number of objects in the field. It may also be pos-

sible, on some machines, to count objects within

objects (e.g., the number of vacuoles within a cell

membrane).

0042 These primary measurements are often combined

to create additional measurement parameters such as:

0043 Equivalent circle diameter: The diameter of a circle

having the same area as the binary image of the

object. Used as a size measurement that is insensi-

tive to the orientation of the object.

0044 Form factors: The dimensionless ratios of primary

measurements whose magnitude gives an indica-

tion of the morphology of an object. Dimensionless

ratios are used to give values that are independent

of size, e.g.:

Circularity ¼ 4p area=perimeter

A perfect circle will have a circularity value of one.

All others shapes will give a lower value.

0045Length: Usually taken to be the longest Feret’s

diameter.

0046Breadth: This can be defined in a number of ways.

For example, the shortest Feret’s diameter may be

used or it may be estimated by dividing the area by

the largest Feret’s diameter. There are other ways of

defining breadth; the choice will depend on the

application.

Three-Dimensional Imaging

0047One of the restrictions of image analysis has been the

need to work on two-dimensional sections. Volumetric

properties have to be derived from stereological com-

putations. Whilst this approach is sufficient for most

work, it does have significant drawbacks: care must be

taken with anisotropic materials to avoid biasing the

data and the sectioning process may mechanically dis-

tort the material. Furthermore, it is necessary to fix and

embed soft or friable material specimens in order to

section them. This may lead to further structural

changes in the material, thus rendering any measure-

ments of dubious relevance to its original state.

0048There is a number of methods for imaging beneath

the surface of structures. Confocal microscopy,

magnetic resonance imaging, and microfocus X-ray

tomography can be used to provide series of two-

dimensional, virtual sections. Many image analysis

packages have software specifically designed to

reconstruct three-dimensional images from a series of

images representing two-dimensional slices generated

by these methods (Figure 3). The reconstructed images

can be used to create further virtual sections in any

direction or to make direct volumetric measurements.

fig0003 Figure 3 (see color plate 110) Reconstruction of puffed wheat from X-ray tomographic sections. Left, reconstruction with cut-out

to show the internal structure; right, reconstruction of the voids inside the puffed wheat. Courtesy of Dr Paul Jenneson, Surrey

University, UK.

MICROSCOPY/Image Analysis 3939

Applications

0049 The potential for applications of image analysis is

enormous – anywhere where rapid, objective, quanti-

tative measurements are required. Particle size and

shape analysis can be used to monitor comminution

or agglomeration processes. The technique has also

been used to measure the air bubble size distribution

in aerated products such as icecream, chocolate, and

batters. Where objects have well-defined morpholo-

gies, shape discrimination can be used to count

several components simultaneously, e.g., counting

the population densities of lactobacilli and strepto-

cocci in yogurt.

The Future

0050 The development of high-resolution scanners for

computer document archiving and the advent of

high-definition TV technology has provided a

number of high-resolution image input devices. This,

coupled with the relentless pursuit of even cheaper

and faster computer memory chips, will provide the

image analyst with access to much higher-resolution

images; 1024 1024 pixel image matrices are already

commonplace and larger formats will follow.

0051 The latest image analyzers use standard computer

components. The only specialized hardware is the

frame-grabber board. New computer interfaces are

being developed that will allow the fast transfer of

data directly into the computer from a digital camera.

This will reduce the cost of the equipment further

and will make image analysis more available to cost-

sensitive applications.

0052Three-dimensional analysis is already an option on

many image analyzers. Its use is restricted due to the

huge amount of memory required and the high

demands made on the processor. These restrictions

will become less relevant and this type of analysis will

become more common in the future.

0053Most electron microscopes and all of the newer

forms of microscopy (e.g., scanning tunneling, atomic

force, and confocal) are controlled via an interface to

a personal computer. These provide an ideal platform

for image processing and analysis.

See also: Microscopy: Scanning Electron Microscopy;

Transmission Electron Microscopy

Further Reading

Gonzalez RC and Wintz P (1977) Digital Image Processing.

Reading, MA: Addison Wesley.

Russ JC (1990) Computer Assisted Microscopy. New York:

Plenum Press.

Serra J (1982) Image Analysis and Mathematical Morph-

ology. London: Academic Press.

Microwave Cooking See Cooking: Domestic Techniques; Domestic Use of Microwave Ovens

MIGRAINE AND DIET

M R Costa and M B A Glo

ria, Federal University of

Minas Gerais, Belo Horizonte, Minas Gerais, Brazil

Migraine Headache

0001 ‘Migraine’ is a French derivative of the Greek word

for hemicrania, or half a head. The problem of mi-

graine is of long standing. Aretaeus of Cappadocia, in

the second century ad, defined migraine as ‘a parox-

ysmal headache disorder that occurred on one side

of the head and then the other, recurred at regular

intervals, associated with vomiting and photophobia,

and ameliorated by dark surroungings.’

0002Throughout the ages, migraine has been a problem

to patients and an enigma to physicians. Migraine is a

public health problem of enormous scope that has an

impact on both individual sufferers and society. Sev-

eral studies have indicated that 82–85% of individ-

uals have some headache-related disability. About

one-third of them are severely disabled or need bed

rest. In addition, many live in fear, knowing that at

any time an attack could disrupt their ability to work,

care for their families, or meet social obligations.

With regard to the impact on society, the costs in

3940 MIGRAINE AND DIET

human terms are overwhelming. The totals, in terms

of lost production, medication, doctors’ and hospital

fees are huge.

0003 Migraine may begin in childhood and usually starts

before the age of 40. Its prevalence in childhood is

about equal among girls and boys, but at puberty

the two groups diverge sharply and females begin to

outnumber males. Frequency tends to decrease with

advancing age, and up to three-fourths of women

note relief after menopause.

0004 It is believed that sufferers from migraines may

have an inherited predisposition for this disorder. It

has been reported that 50–70% of patients with mi-

graine have a positive family history. The precise

mode of inheritance is not known, but current

thought is that this predisposition is an autosomal

dominant trait characterized by incomplete pene-

trance.

Symptoms of a Migraine Headache

0005 The symptoms most frequently associated with mi-

graine headache are indicated in Table 1. Migraine is

characterized by throbbing head pain. Most of the

time, it is unilateral in adults; pain may occur on

either side of the head but rarely changes sides during

an attack. In children, migraine is ordinarily bilateral.

It is often accompanied by nausea and sensitivity

to light and/or sound. Less frequently noted are

abdominal bloating and diarrhea, pallor, sweating,

chills, and cold extremities. Every individual is

different and may experience some symptoms and

not others, or in various combinations. Intensity

varies from one attack to another.

0006The frequency of migraine ranges from every few

days to less than once a year. However, most patients

experience one to four attacks a month. These gener-

ally last from 4 h to 2 days, but severe episodes of

prolonged migraine may persist for weeks. Proper

management can significantly reduce the impact of

this debilitating chronic disorder and improve both

functioning and quality of life.

Pathophysiology of Migraine

0007Migraine is a vascular headache believed to be caused

by blood flow changes and certain chemical changes

in the brain leading to a cascade of events, including

constriction of arteries supplying blood to the brain

and the release of certain brain chemicals.

0008Migraine has a number of forms, some of which

may be confused with other diseases. It can be classi-

fied (Table 2) as migraine with aura (classic) and

without aura (common).

0009The exact cause of migraine is uncertain,

although various theories are being studied. The

neurogenic theory emphasizes the significance of

active substances acting upon the central nervous

system and promoting intracerebral vasoconstriction

and extracranial vasodilatation. According to many

researchers, migraine is due to the vulnerability of the

nervous system to sudden changes in either the body

or the surrounding environment. They believe that

migraine sufferers have inherited a more sensitive

nervous system response than those without mi-

graine. During a migraine attack, changes in brain

activity produce inflamed blood vessels and nerves

around the brain.

0010Another theory is that migraine represents a dis-

order of cerebrovascular regulation. In the typical

migraine sequence, a trigger precipitates a cascade

of events that includes release, activation, and pro-

duction of active substances. When these substances

tbl0001 Table 1 Symptoms associated with migraine headache

Symptoms of migraine headache

Moderate to severe headache pain

Duration of 4–48 h, sometimes longer

Throbbing pain located on one side of the head

Nausea

Vomiting

Sensitivity to light, sound, or odors

Pain becomes worse when moving

tbl0002 Table 2 Symptoms and signs of different forms of migraine

Types ofmigraine Characteristics

Migraine with aura (classic) Neurologic symptoms occur 15–60 min before headache: scotomas or blind spots

are common; photopsia and teichopsia or fortification spectra occur frequently;

diplopia, macropsia, and metamorphopsia are less common

Headache begins as neurologic symptoms diminish

Migraine without aura (common or nonclassic) Occurs in 85% of all patients

No neurologic aura precedes the headache

Premonitory disturbances of mood or appetite (craving for sweets, euphoria,

fatigue, yawning) can happen 1–3 days before an attack

Bilateral or generalized pain

MIGRAINE AND DIET 3941

concentrate in the tissue, vascular and nervous phe-

nomena are enhanced. A fact of paramount import-

ance is that some of these substances potentiate

one another. In patients with regular attacks, each

episode of migraine is followed by a phase of immun-

ity and then by one of increasing reactivity, which

can be divided into two stages. The first is called

adrenergic, and is characterized by relatively mild

symptoms, such as no pain, vasoconstriction, slight

excitation, and sensory manifestations. In the second

one, there is intense pain, marked vasodilatation, and

pronounced psychic depression. During the reactive

phase, migraine episodes may be triggered by stimuli

that promote vasoconstriction (e.g., cold, anxiety) as

well as vasodilatation (e.g., heat, alcohol, nitrites).

Triggers or Provokers of Migraine

0011 Certain factors can provoke or trigger attacks more

frequently in those susceptible to migraine headache.

Some possible triggers include biochemical or envir-

onmental factors; hormonal changes in women; too

much or too little sleep; stress (feeling overly tired);

pharmacological precipitants; and diet. According to

many studies (Table 3), the most commonly cited

triggers are stress, change in sleep patterns, menstrual

cycle, hunger, and diet or specific foods. Not all mi-

graines have the same provoking factors, nor do all

these factors necessarily provoke an attack. Migraine

can also arise from a number of factors acting in

unison.

Biochemical Factors

0012 It has been postulated that there is a biochemical

defect in migraine sufferers, especially with respect

to their ability to metabolize amines and phenols

when compared to controls. The monoamine oxi-

dases (MAO) act as a barrier against the absorption

of vasoactive substances contained in foods. MAO

defect can prevent amine metabolism and, therefore,

cause accumulation of monoamines. There could also

be a deficiency of phenolsulfotransferase P, which is

specific for the detoxification of dietary phenols.

Environmental Triggers

0013Environmental triggers of migraine headache include

weather or temperature changes, exposure to bright

lights (glare from snow or fluorescent lights, com-

puter screens), loud noises, and strong odors. High

altitudes or the atmosphere of a pressurized cabin in a

commercial airliner may also lead to migraine attacks

or similar vascular headaches.

Menstrual Cycle

0014Many women with migraine have attacks linked to their

menstrual cycles. Menstrual migraine occurs mainly

between 2 days prior and 3 days following the onset of

flow. It is usually related to changing levels of menstrual

hormones during menstruation, puerperium, hysterec-

tomy, and pill-taking. Migraine prevalence peaks in

womenintheirearly40sandthendeclinessteadily

with age. Migraine may worsen in early pregnancy, but

it generally improves in later pregnancy. Although some

women find that their migraines worsen when taking

oral contraceptives, the majority experience no change

or improvement in their headache pattern.

Pharmacological Agents

0015Some pharmacological agents used to treat hyperten-

sion and coronary artery disease may increase mi-

graine frequency. Reserpine may increase migraine

frequency and exacerbate depression in some pa-

tients. Amyl nitrite, used for treatment of angina

pectoris, may adversely affect the frequency of mi-

graine. Migraine was observed after beginning 25 000

IU daily supplementation of vitamin A.

0016Headache potential is also increased in patients

treated with MAO inhibitor drugs, such as phenyl-

zine, isocarboxazid, pargyline and tranylcypromine

for depression; iproniazid and isoniazid for tubercu-

losis; procarbazine hydrochloric for cancer; and

furazolidone for antimicrobial use.

Emotional Stress and Anxiety

0017Stress is the most common migraine trigger. Emo-

tional stress or response to life’s daily pressure can

trigger migraine attack in some individuals. Stress

can cause alteration in adrenal hormone level. Stress,

fear, and anger may liberate vasoactive substances

from tissue stores.

0018Depression has been associated with migraine in both

prospective and retrospective epidemiological studies,

and psychological distress has been shown to be respon-

sive to cognitive-behavioral headache interventions.

Sleep Patterns

0019An extra 1–2 h sleep can be enough to initiate a

headache. The mechanism involved is not understood

tbl0003 Table 3 Major precipitating factors in migraine

Precipitating factors Frequency (%)

Heredity 66

Stress, anxiety, and worry 43–88

Change in sleep patterns 77

Menstrual cycle 39–65

Oral contraceptive 31

Bright light 30

Diet or specific foods 13–74

Fasting 7–77

3942 MIGRAINE AND DIET

but carbon dioxide retention with vasodilatation or

delayed breakfast provide two possibilities. However,

lack of sleep for social or work reasons, as well as

psychological factors, can be a migraine precipitant.

Hunger and Hypoglycemia

0020 Blood glucose levels have important effects on the tone

of cerebral blood vessels. Marked cerebral vasodilata-

tion may occur if blood glucose levels are reduced by

insulin or other causes. Therefore, hunger headache

can occur during fasting or food deprivation. It is

provoked after stopping food intake for 5 h during

the day or 13 h overnight. This condition may also be

responsible for the weekend headache that develops

when a person sleeps late and delays the morning

meal. Since blood glucose levels do not fall below

normal levels, it cannot be called hypoglycemia. How-

ever, hypoglycemia has also been associated with

headache. It is a prominent symptom in a patient

entering insulin shock. In addition, the reactive hypo-

glycemia that occurs after ingestion of a large simple-

carbohydrate meal may also precipitate a similar

headache in susceptible persons. Therefore, in order

to reduce the number and severity of migraines, a

patient should not skip meals, should have three or

more well balanced meals a day and should avoid

overabundance of carbohydrates at any single meal.

Cold Drinks or Icecream

0021 Swallowing cold drinks or icecream can quickly cause

a brief frontal icecream headache lasting a few min-

utes. The pain occurs either in the palate and throat or

the forehead, temple, or ear. The cause is believed to

be the sudden cooling of the oral pharynx. Icecream

headache is an excessive vascular reaction to cold and

a manifestation of erratic vasomotor regulation in

patients with migraine.

Allergy

0022 There is a very long medical history which attempts

to relate migraine to allergy. However, the view of

many researchers in the field of migraine appears to

be that dietary migraine is not an allergic phenom-

enon. Recently, it has become evident that the

immune system might be affected in migraine. Foods

such as milk, wheat, eggs, and meat are usually in-

volved. In patients with migraine, these foods induce

formation of immune complexes which can release

active substances. They may also react with mast

cells, releasing histamine and serotonin.

Influence of Diet

0023 Studies on the influence of diet on migraine are diffi-

cult to evaluate. A diet trial cannot be performed in a

double-blind manner. Moreover, it is hard to reduce

selectively one or a couple of components in the diet

and yet preserve the original proportions of nutrients.

Diet trials are also difficult to carry out with regard to

compliance and practical inconvenience for the

patients.

0024Although the relationship between diet and mi-

graine has been a source of controversy for many

years, certain relationships have become clear. Some

specific foods, alcoholic beverages, and some food

additives are possible triggers of diet-precipitated

migraine, even though sensitivity varies with each

patient.

0025Chocolate, cheese, and citrus fruits are the most

frequently cited provoking factors to which patients

ascribe their dietary migraine (Table 4). Fatty fried

foods, vegetables, tea and coffee, pork, and seafood

have also been cited. In 490 migraine patients

attending the Princess Margaret Migraine Clinic at

Charing Cross Hospital in 1984, 19% cited choc-

olate, 18% cheese, 11% citrus fruit, 29% reported

alcohol, 7% coffee, 3% pork and dairy products, and

2% eggs as precipitants of their headaches. In another

study with 429 migraine patients attending the same

clinic from 1989 to 1991, 17% reported that head-

aches could be precipitated by cheese and chocolate,

and nearly always by both; 18% reported sensitivity

to all alcoholic drinks, while another 12% were sen-

sitive to red wine but not to white wine; and 28%

reported that beer would precipitate headaches.

0026Food-induced headaches are mediated by chemical

constituents common to these foods (Table 5). Some

are naturally occurring components, including tyram-

ine, phenylethylamine, serotonin and other vaso-

active amines, alcohol, flavonoid, caffeine, and

copper. Additives added to foods, such as sodium

nitrite and monosodium glutamate (MSG), may also

elicit migraine headache. Some other food or food

tbl0004Table 4 Major precipitating foods in migraine

Precipitating factors Patients withdietary migraine (%)

Chocolate 17–75

Alcohol 11–67

Red wine 12

Beer 28

Cheese 17–50

Citrus fruits 11–30

Nuts 12–17

Fatty fried foods 18–19

Coffee 7–15

Ham 4

Pork 3–15

Dairy products 3–9

Seafood 6–10

Onion 18

MIGRAINE AND DIET 3943