Review of Medical Physiology - William F. Ganong 2002

Appendix

MAIN SOURCES

When studying physiology, it is advisable to rely on comprehensive fundamental textbooks. Among the best published over the past decade are:

Berne RM, Levy MN (editors): Physiology, 3rd ed. Mosby, 1993.

Guyton AC, Hall JE: Textbook of Medical Physiology, 9th ed. Saunders, 1996.

Johnson LR (editor): Essential Medical Physiology, Raven-Press, 1992.

West JB (editor): Best and Taylor's Physiological Basis of Medical Practice, 12th ed. Williams & Wilkins, 1990.

McPhee, Lingappa, Ganong, and Lange recently published the third edition of their Pathophysiology textbook for medical students:

McPhee SJ et al: Pathophysiology of Disease: An Introduction to Clinical Medicine, 3rd ed. McGraw-Hill, 2000.

Other noteworthy textbooks on pathophysiology include:

Frohlich ED (editor): Pathophysiology, 3rd ed. Lippincott, 1984.

Sodeman WA Jr, Sodeman WA: Pathologic Physiology, 6^th ed. Saunders, 1979.

A classic textbook on anatomy is:

Bannister LH et al (editors): Gray's Anatomy: The Anatomical Basis of Medicine and Surgery, 38th ed. Churchill Livingstone, 1995.

Modern diagnostic imaging techniques, which have become a cornerstone of contemporary physiology and medicine, are covered in:

Haaga JR, Alfidi RJ (editors): Computed Tomography of the Whole Body, 2nd ed. Mosby, 1988.

Von Schulthess GR: Clinical Positron Emission Tomography. Lippincott Williams & Wilkins, 1999.

Excellent reviews of current research across various fields of physiology can be found in the News, Reviews, and Perspectives sections of Nature and Science. Feature articles and general reviews regularly appear in the New England Journal of Medicine; these include Overview articles on recent advances in physiology and biochemistry that provide up-to-date information for practicing clinicians. Particularly valuable are the review series published in Physiological Reviews, Pharmacological Reviews, Annual Review of Physiology, and other journals in these series.

The recently published Handbook of Physiology (Oxford University Press, New York) comprises individual volumes covering all areas of physiology. While scientifically valuable, these chapters are exceptionally detailed.

NORMAL VALUES AND STATISTICAL Data analysis

Normal physiological ranges for several commonly measured plasma components are provided in the tables on the inside back cover of this book. A global transition toward a unified, standardized nomenclature based on the SI system is currently underway. The system is built upon seven fundamental physical quantities (see Table 1). Derived units are listed in Table 2, and standard prefixes are given in Table 3. Certain challenges regarding The Use of these units—such as expressing enzyme activity units—are rarely discussed in medical literature. In this book, values within the text are presented in traditional units, but in most cases are accompanied by their SI equivalents.

The accuracy of laboratory Methods varies. When evaluating any individual measurement, it is essential to account for potential sources of error. For chemical assays of Body Fluids, these include errors associated with sample collection and the analytical chemical method itself. Furthermore, even when using the most precise methods, results obtained from different normal individuals will vary due to so-called biological variation. This Variability arises because any living system—whether an Organism or a tissue—possesses numerous intrinsic factors that influence a specific measurement. Variables such as age, sex, time of measurement, and time since the last meal must therefore be taken into consideration.

Normal reference ranges for any physiological or clinical measurement can be determined using standard statistical analysis, provided the measurements are performed on an adequate sample from a healthy population (preferably over 20 subjects). It is important to know not only the mean value for such a sample, but also the range of individual values deviating from the mean.

Table 1. Main SI units

The mean value (arithmetic mean, M) of a dataset is calculated using the formula

Class="center">

where X represents individual values; n is the number of individual measurements in a series.

The mean deviation is the average of the deviations of each data point from the arithmetic mean. From a mathematical standpoint, deviations are best determined using the geometric mean of the deviations from M, which is termed the sample standard deviation (s):

For mathematical reasons, n-1 is used instead of n. The value of s may differ from the population standard deviation, denoted by σ. However, if the sample is truly representative, s and σ will be comparable.

Another commonly used measure of data dispersion is the standard error of the mean (SEM):

The SEM reflects the reliability of the sample mean as an estimate of the true population mean from which the sample was drawn.

A frequency distribution curve can be constructed from individual population values by plotting the frequency with which any given value occurs in a series relative to other values. If the test group is homogeneous, the distribution curve is symmetrical (Fig. 1) with the highest frequency at the mean value, and the width of the curve depends on σ (normal distribution curve). Within an ideal normal distribution curve, the percentage of observations falling within various limits is presented in Table 4. The mean and s in a representative sample approximate the mean and σ of the entire population. Therefore, based on the sample mean and s, one can predict the probability that any individual value in the general population is normal. For example, if the difference between such a value and the mean is 1.96s, the probability of it being normal is 1 in 20 (5 in 100). Conversely, the probability that the value is abnormal is 19 in 20.

Table 2. SI Derived Units

Statistical analysis is also used to determine the difference between two mean values. In physiological and clinical research, measurements are frequently performed on a group of animals or patients receiving a Treatment. These measurements are compared with identical measurements made on a control group that ideally has undergone identical conditions and treatment, except that no treatment was administered. If the mean for the treatment group differs from the corresponding control mean, the question arises whether this difference is the result of the treatment or of data dispersion. The probability that the difference represents data dispersion can be determined in many cases using Student's t-test. The t statistic is The ratio of the difference between the means

Table 3. Standard Prefixes¹

¹ These prefixes are used for SI units and others. For example, a micrometer (μm) is 10^-6 meter (also called a micron); a picoliter (pL) is 10^-12 liter; a kilogram (kg) is 10³ grams

Fig. 1. Normal distribution curve (frequency distribution curve of values for a homogeneous population).

of two series (M_a and M_b) to the standard error of these means. The formula used for calculations is

where n_a and n_b are the number of individual measurements in series a and b, respectively. If n_a = n_b, the formula for determining t is simplified:

The larger the value of t, the lower the probability that the difference represents random data dispersion. This probability decreases as the Sample size n in each group increases, because a greater number of measurements reduces measurement error. The mathematical expression of probability P for any value of t at various values of n can be found in tables provided in most statistics textbooks. The P value is a fraction representing the probability that the difference between two means is due to data dispersion. For example, if P equals 0.10, the probability that the difference is due to chance dispersion is 10% (1 chance in 10). A P value < 0.001 indicates that the probability of the difference arising from data dispersion is less than 1 in 1000. If P < 0.05, most researchers consider the difference "statistically significant"; that is, the difference is considered to result from a specific factor rather than random chance. The use of the t-test is only appropriate for comparing two groups. In experiments involving more than two groups, a systematic error will occur, leading to an inflated probability value. In such cases, analysis of variance (ANOVA) is applied. This and other techniques are described in statistics textbooks.

Table 4. Percentage of population values falling within various Structure/21.html">Limits of the normal distribution curve

The elementary methods mentioned, along with many others suitable for statistical analysis in research laboratories and clinics, ensure the determination of objective mean values. Statistical significance is not arbitrary for a physiological mean, and the reverse is sometimes true; however, replacing subjective impressions with the results of statistical analysis is an essential goal in medicine.

Recommended books on statistics include:

Dawson-Saunders B, TrappRG: Basic and Clinical Biostatistics, 2nd ed. Appleton & Lange, 1994.

Rosner B: Fundamentals of Biostatistics. Duxbury, 1982.

The SI for the Health Professions: World Health Organization, 1977.

Zar JH: Biostatistical Analysis. Prentice-Hall, 1974.

Standard Respiratory Symbols

(see Handbook of Physiology, Section 3: The Respiratory system. American Physiological Society, 1986)

Metric, US, and British Measurement Equivalents

(values rounded to two decimal places)

Length

1 kilometer = 0.62 miles

1 mile = 5280 feet = 1.62 kilometers

1 meter = 39.37 inches

1 inch = 1/12 FOOT = 2.54 centimeters

Volume

1 liter = 1.06 US quarts

1 US quart = 32 ounces = 1/4 US gallon = 0.95 liters

1 milliliter = 0.03 ounces

1 ounce = 29.57 milliliters

1 US gallon = 0.83 Imperial gallons

Mass

1 kilogram = 2.2 pounds (avoirdupois) = 2.68 pounds (apothecaries')

1 pound (avoirdupois) = 16 ounces = 453.60 grams

1 grain = 65 milligrams Energy

1 kilogram-meter = 7.25 foot-pounds

1 foot-pound = 0.14 kilogram-meters

Temperature

To convert degrees Celsius to degrees Fahrenheit, multiply by 9/5 and add 32

To convert degrees Fahrenheit to degrees Celsius, subtract 32 and multiply by 5/9

Greek Alphabet