Population differences in cutaneous methacholine reactivity and circulating IgE concentrations

Population differences in cutaneous methacholine reactivity and circulating IgE concentrations C. Edward Buckley III, M.D., James W. Larrick, M.D., Ph.D.,* and Jonathan E. Kaplan, M.D. Durham, N. C., Palo Alto, Calif.., and Atlanta,

Ga.

We evaluated the incidence of allergic and vasomotor symptoms, serum IgE concentrations, and the cutaneous responses to allergens an&or methacholine in 229 Waorani Indians residing at 300 m altitude near the headwaters of the Amazon River, 39 Tibetans residing at 4000 m in the Himalayas, and 84 healthy subjects residing at IS0 m in the Piedmont region of North Carolina. The Waorani Indians had a high level of intestinal parasitism, an intermediate level of parasitism occurs in Tibetans, and parasitism is rare in the control population. One Waorani Indian (
Immediate hypersensitivity diseases are mediated by IgE antibodies to allergens.’ In spite of this relationship, the severity of allergic symptoms cannot be reliably predicted by variation in serum IgE concentrations or by the activity of IgE antibodies.*, 3 The relationship between IgE and allergic symptoms could be obscured by the autonomic nervous system alterations associated with allergic disorders. Evidence of autonomic dysfunction in allergic patients can be deFrom the Departments of Medicine and Microbiology-Immunology, Duke University, Durham, N. C., Cetus Immune Research Laboratories, Palo Alto, Calif., and University of New Mexico School of Medicine, Alburquerque, N. M. *Supported by Library of Medicine Grant LI-000422 and National Institutes of Health Training Grant GM-07171 (Medical Scientist Training Program). Received for publication Sept. 7, 1984. Accepted for publication March 5, 1985. Reprint requests: C. Edward Buckley III, M.D., P. 0. Box 3804, Duke University Medical Center, Durham, NC 27710.

tected in the skin and many other tissues.4 The relative importance of immune and autonomic factors in the pathogenesis of allergic disease is not known. Before evaluating both factors in allergic patients in whom disease might provide an additional source of variability, it appeared prudent to evaluate the variation of each factor in different populations where genetic, ecologic, climatic, and cultural differences might alter their adaptive importance. An opportunity to make this comparison arose during studies of the Waorani Indians who reside in the rain forest at the headwaters of the Amazon River in Ecuador and Tibetans who reside in the Himalayas of Nepal. Ecologic and cultural factors predispose to an appreciable level of intestinal parasite infestation and elevated serum IgE concentrations in both populations. Serum IgE concentrations and the cutaneous reactions to methacholine were measured in each population and compared with the same observations in inhabitants of the central Piedmont region of North 847

TABLE I. Populations sampled for allergic factors associated with allergic disease ----.___

and vasomotor

symptoms

and for immunophysiologic:

-____ Altitude

Population

--

Waorani Indians Tibetan subject\ North Carolina subjects ___-

Location

Ecuador, Amazon headwater\ Nepal. Himalayan mountains Carolina Piedmont --__

Carolina. The large differences in serum IgE concentrations and cutaneous methacholine reactivity observed suggest possible alternative explanations for the absence of allergy’-’ in primitive subtropical populations. These observations provide an important perspective of similar observations in allergic patients.

METHODS Subjects and clinicaJ studies The differences in the geographic location. altitude, climate. and relative prevalence of intestinal parasites in the three subject groups studied are summarized in Table 1. The Waorani Indian and Tibetan subjects were recruited from among those attending medical and dental clinics focused on multidisciplinary studies of health and disease in primitive populations. Trained linguists were available on both expeditions. The health of each subject was clinically evaluated, and those subjects who were clinically well by history and physical examination were included in this study. By clinically well we mean those subjects who were free of symptoms or signs of disease and capable of physical activities and endurance equivalent to or greater than appropriate expedition member control subjects. Each subject’s past history of symptoms of immediate hypersensitivity (anaphylaxis. asthma, allergic rhinitis, angioedema. or urticaria) and vasomotor symptoms (temperature change-induced nasal congestion or rhinorrhea, light-induced sncezing. postural presyncope. crythema fugax, or Raynaud’s phenomena) was obtained as a part of their medical evaluation. Similar clinical observations were available from 84 previously reported white and black IS- to 35year-old apparently healthy volunteers who resided in the Piedmont region of North Carolina.” The Waorani are American Indians who inhabit the Amazon rain forest south of the Rio Napa in eastern Ecuador.” The Waorani were studied during the winter months of July and August in the southern hemisphere. The equatorial climate remains hot throughout the year. and seasonal temperature tluctuations are so small that the Waorani do not measure time in relation to seasonal change. The ages of the Waorani were obtained as clinical estimates based on genealogies available from prior expeditions. The Waorani have a hunter-agricultural economy ( Fig. I ) and eat a variety of game animals (monkeys. tapirs, rodents, and birds) and domesticated plants (bananas, maniac, and peanuts). Settlements are composed of one to several extended family groups, each inhabiting a palm-thatched A-frame long

(ml

Climate

Parasitism ---.

.m 1 4lnK~ 150

H(ll (‘i)ld Tcmpcratc

H1gli \I ,rciLm!i. ! _ia\‘\

house. Gardens are planted in several locations at staggered times throughout the year in order to ensurc .I ~~rntinuous supply of food. An extended family group vvi!l ~n’tcnmove several times a year to harvest its gardens and to exploit ii different hunting territory. The Waorani have angrizong dnd Phalag in the upper Kali Gandaki River bakin in central Nepal behind the Dhaulagiri massif. Although seasonal changes occur. the climate remains cool throughout the year. The Tibetans were studied during the fall. Like the Waorani. the people of these villages are Mongoloid IV ,brrgin. The Tibetans have cultural, religious. and linguistic ties to Tibet (Fig. 2). Their subsistence culture is classified 4s agropa\toralist. During the warmer months of the !c,:r. families reside in highland settlements and herd sheep. g~;rt\. cattlc yak, and yak-cattle hybrids (dzo) in the high nn~untain I!,& leys. During the remainder of the year. the Il‘(bet,rn\ iivc m small adobe villages and worh ticids of buckwbcat. b,uLr\. wheat. millet. and radish. Although this wa\ IIO! Mt:lI cbaracterized, a modest prevalence of intestinal parasitism cxis~> among Tibetans. ”

Skin tests A solid bibarbed prick needle was used to mducc a curcular scratch and deposit a microdrop of allergen or agonist solution on the acetone-cleansed skin. The physical attributes and variability of these needles, preparation of methacholine solutions, and the reliability of the skin test method has been previously described.” Sufficient space was alLowed between test sites to avoid confluent flare rcspomes The longest diameter and a bisecting orthogonal diameter ot’2ach flare and wheal response were measured to the nearest tnillimeter after IS to 20 minutes. The average of the two diameters was recorded. The maximum methacholine concentration used for skin testing was 1777 mmol/L.. and serial twofold dilutions contained 888, 444. or 232 mmol/L ot methacholine. All skin tests were done on the volar surface\ of the forearms or over the upper back: prior studies with pharmacologic agonists and allergens suggest cyuivalentsized reactions are elicited at both sites. The neat allergenic extracts (Greer Laboratories. Lenoir.

VOLUME NUMBER

76 6

Cutaneous

methacholine

reactivity

and IgE concentrations

849

FIG. 2. A group of Tibetan women and children who reside in the cold climates at 4000 m altitude in the mountain valleys of the Himalayas. Note the mountain peak in the mist in the background and the clothing used to protect against the cold. FIG. 1. The Waorani Indians reside at 300 m altitude in the rain forest in Ecuador near the headwaters of the Amazon River. This hunter is carrying a blowgun and dressed appropriately for daily activities in his environment.

N. C.) used to skin test the Waorani were selected from among those generally available on the basis of descriptions of the flora of the rain forest where the Waorani reside. The extracts included preparations of timothy (Phleum prutense). rye grass (Lolium perenne), Bermuda (Cywdon dacr$on). Johnson (Sorghum halepense), and smooth bromegrass (Bromus inermis) grasses. alfalfa (Medicago x&a), short ragweed (Ambrosia artemisiifolia). lamb’s-quarters (Chenopodium album). rough pigweed (Amarunthus retro~l~.rus). Russian thistle (Salsnla k/i) weeds. chrysanthemum (Chrysanthemum sp) and marigold (Togetespatula) Howers. and white oak (Quercus alba), mango (Mangfem irldica). Australian pine (Casuarina equisettfolia). mesquite (Prosupis ~kmdulosa). eucalyptus (Euca!\ptus glohulu.s), and date palm (Phoenk dactylifera) trees. Because of the lack of information about local flora and the small number of subjects studied, allergen skin tests were not done among the Tibetans. The allergens used among the residents of the North Carolina region have been previously described.”

Circulating

IgE concentrations

Heparinized blood was obtained by venous phlebotomy from the Waorani and Tibetan subjects. Plasma was separated by gravity, withdrawn, and made to 0.02% w/v with

sodium azide. The plasma samples obtained in Ecuador were frozen immediately in liquid nitrogen. The plasma samples obtained in Nepal were maintained at
Evaluation

of observations

The results of skin tests with three or more concentrations of methacholine were evaluated in each subject by a procedure previously described’ in order to obtain a quantitative bivariate parameterization of methacholine responsiveness (8) and sensitivity (C) from the flare responses provoked by the different concentrations of methacholine. The closeness of the fit of the bivariate parameterization of the sigmoid dose-response curve in each subject was evaluated by

850

Buckley

et al.

TABLE II. Allergic factors important

and vasomotor symptoms in allergic disease

in three populations

Waorani Indians Tibetan subjects North Carolina subjects

Symptoms

. .Prevalence

I Allergic 6 Vasomotor 17 Allergic IX Vasomotor ?9 Combined

229 39 84

TABLE Ill. Comparison of skin test reactions and United States subjects*

for immunophysioloyr(.. _-_-

No. of subjects

Population

sampled

to extracts

of allergenic

pollens

in Waorani

Indians

Skin tests

Populrtion

Waorani Indiansi’ North Carolina subjects*

No. of subjects

194 84

Positive no.

40 151

%

(1.3) (11.7)

Negative no.

2870 1193

-%

(98.6, (88.8)

_I_..-. Total no.

3910 1344

*The prevalenceof reactionsin the North Carolina subjectswas higher than in the Waoranisubjects(p s .OOOl:chi-squared := iXJ.486i. tThe saline control was negative and positive reactions (23 mm) were identified with 15 of 18 skin test allergens. $The saline control was positive in approximately half the subjects studied, and positive reactions (a-4 mm) were identitied with 16 oi Ih allergens. calculating a chi-squaredlike parameter similar to that described by Smith.lS Values of l? are normally distributed. Values of c are log normally distributed and were transformed before use as a dependent variable in analyses of variance. Averages and the standard error of the mean were compared by use of an F ratio with the appropriate number of degrees of freedom by methods described by Mather.‘” Observed differences were interpreted as significant when the null hypothesis could be rejected with reasonable confidence (p s 0.05).

RESULTS Symptoms

and allergen

skin test responses

Table II summarizes the subject’s past history of clinical symptoms. Ail subjects were clinically well as evidenced by their medical history and a physical examination. The physical strength and endurance of the adult subjects in the primitive populations equaled or exceeded that of expedition member control subjects. Within the limitations imposed by the symbolic content of their language, the past prevalence of allergic and vasomotor symptoms among the Waorani was < 1%. Among the Tibetans, 15% reported past symptoms consistent with vasomotor disorders. In contrast, 88% of the control subjects from North Carolina reported a past history of subtle symptoms consistent with allergic (20%)) vasomotor (2 1%), or both (46%) phenomena. Table III summarizes the results of allergen skin

tests. Allergen skin tests were done with I8 common pollen allergens and a saline control in 194 Waorani Indian subjects. Saline control reactions were d I mm. Positive skin test responses is-3 mm) were detected among the Waorani with 15 of the allergens. Two Waorani reacted to nine pollen allergens; one to eight 1 three. and two allergens, respectively; and nine reacted to one allergen. Seven Waorani reacted to allergenic extracts from white oak; five to Australian pine: four to short ragweed: three to rye grass. Russian thistle. and eucalyptus, respectively; twe reacted to chrysanthemum. marigold, date palm. and mango. respectively; and one to lamb’s-quarter, mesquite. timothy grass, pigweed, and alfalfa? respectively. Considering only these 15 positive allergens, 30 positive tests (a-3 mm) were detected in 2910 ( 15 allergens times 194 subjects) allergen skin tests. The 84 subjects studied in North Carolina were skin tested with 16 common pollen allergens. Unlike the Waorani, reactivity to a saline control was detected in the North Carolina subjects, and a slightly larger Aare reaction was used to identify a positive skin test. One hundred fifty one positive responses 1~~4 mm) were detected among 1344 allergen skin tests in this subject sample. Because of the lack of specific information about the local flora and the small number of subjects. allergen skin tests were not done in the Tibetans. Even with the conservative adjustment from 18 to 15 ai-

VOLUME NUMBER

Cutaneous methacholine reactivity and IgE concentrations

76 6

IV. Comparison three populations

TABLE

of serum IgE concentrations

and cutaneous

reactivity

to methacholine

851

in

Methacholine

Population

No. of subjects

Waorani Indians Tibetan subjects North Carolina subjects

229 39 84

Serum IgE (NJ/ml)

(1) (2) (3) (1) (2) (3) (1) (2) (3)

9,806 9.1908 t ,044s 2,930 7.9826 k .23.59 108 4.6858 L .I968

Responsiveness (flare, mm)

Sensitivity (dose, mmol/L)

Cl.00 -

4.49 t .08

237 5.4696 k.1748 287 6.0222 -e .0749

0.0001

0.0929

2.06 k.12

Comparisons(ps) Waorani vs. Tibetans Waoranivs. N.C. subjects 0.0001

Tibetans vs. N.C. subjects (1) = Geometric mean; (2) = log mean; (3) *A dose-response curve could not be elicited and was taken as the maximum estimate of siveness over all subjects was r = -0.50,

= standard error of the mean. with methacholine in the Waorani Indians; detectable reactions to the agonist were < 1 mm responsiveness. The correlation between serum IgE concentrations and estimates of responp C 0.0001.

lergenic pollen extracts reactive among the Waorani and the use of a slightly larger flare reaction (a4 mm) to identify positive reactions in the North Carolina subjects, the large difference between the 1.3% incidence of positive responses in the Waorani Indians and the 11.2% incidence of positive responses in subjects from North Carolina was highly significant (chisquared = 184.486;~ < 0.0001). Serum IgE concentrations The geometric mean concentration of IgE in serums from Waorani subjects in whom the sex was known was 9503 IU/ml in male Indians (log mean + SEM, 9.1594 + 0.0715, n = 97) and was not significantly different from the 10,100 IU/ml concentration of IgE in female Waorani subjects (9.2202 ? 0.0704, n = 98). When it was possible, the age of Waorani subjects was estimated to the nearest decade of life. In serums from Waorani subjects in whom age estimates were made, the decade of life, number of subjects, geometric mean serum IgE concentrations, and log mean 2 SEM from the first through the fifth decades were, for the first decade, 10,227 IU/ml, 9.2327 L 0.1364, n = 37; second decade, 10,797 IU/ml, 9.2870 + 0.0764, n = 77; third decade, 9760 IU/ml, 9.1861 & 0.0798, n = 39; fourth decade, 7919 IU/ml, 8.9770 + 0.1636, n = 26; fifth decade and older, 7848 IU/ml, 8.9680 ? 0.1442,

n = 19. The overall geometric mean in all 229 subjects was 9806 IU/ml with a log mean and standard error of 9.1908 ? 0.0448. A similar evaluation of age-related differences in IgE was not done in the small number of Tibetans studied. A single stool examination for helminth and protozoan parasites was done in 44 (19.2%) of the 229 Waorani subjects. These stool examinations provided a minimum estimate of parasitism in the population. Parasites were not detected in the stools of nine (20%) of the subjects who had a geometric mean serum IgE concentration of 7609 IU/ml (log mean +- SEM, 8.9371 2 0.3616). This average was not significantly different from 5247 IU/ml geometric mean in three Waorani with only protozoan species in the stool (8.5654 + 0.1164) and the 7244 IU/ml geometric mean in 16 Waorani with only helminth species in the stool (8.8879 & 0.2407). In contrast, 16 Waorani with helminth and protozoan species in the stool had a geometric mean serum IgE concentration of 11,597 IU/ml(9.3585 ? 0.1407), and this average was significantly higher (p G 0.0419) than the pooled average of serum IgE concentrations in subjects with one or the other or no detectable stool parasites. This suggests the high concentrations of IgE in the Waorani reflects the degree of intestinal parasite infestation. The average serum concentrations of IgE varied among the three subject populations. Table IV sum-

852 Buckley et al. marizes the geometric mean. log mean. standard error, and standard deviation of serum IgE concentrations in all subjects. The geometric mean serum IgE concentration in the subjects residing in North Carolina was 108 IUiml. The average serum IgE concentrations among Tibetans was 27-fold higher and averaged 2930 IUiml. The geometric mean serum IgE in the Waorani Indians was 9,806 IUlml, an average that is 91-fold higher than the average observed in subjects from North Carolina. Differences between all three population samples were highly significant. Cutaneous

reactivity

to methacholine

Even though the flora of the rain forest was not very different from selected plants found in the United States and the Waorani exhibited cutaneous reactivity to a few pollen allergens. the possibility exists that the comparison presented in Table III underestimates their cutaneous reactivity. Skin tests with allergenic extracts of local pollens was not possible. Reservations about prior pollen exposure do not apply to skin tests with methacholine, a pharmacologic agonist that activates cholinergic mechanisms in many forms of animal life. Table IV also summarizes the results of an analysis of cutaneous reactions to selected concentrations of methacholine. All skin test reactions to methacholine in the Waorani were equivalent to tests with a saline control (5 1 mm). In contrast, estimates of responsiveness to methacholine in the Tibetans. i.e.. half the height of the sigmoid dose-response curve, averaged a 2.06 t 0.12 mm flare response. Conservatively, this magnitude of responsiveness was at least twofold greater than a 1 mm assumed maximum average among the Waorani. Responsiveness to methacholine among the 84 subjects sampled in North Carolina averaged 4.68 +- 0.19 mm. This value was more than fourfold greater than the assumed maximum response in the Waorani. The differences in responsiveness among all three populations were highly significant. In contrast, the difference in sensitivity, i.e.. the dose of methacholine needed to elicit a flare reaction equivalent to the estimate of responsiveness. between the Tibetans and North Carolina subjects was small and not significant. This suggests the difference in cutaneous methacholine reactivity detected between the two populations is primarily related to differences in responsiveness. Since reliable dose-related changes in the flare reaction were not possible in the Waorani, an estimate of their cutaneous methacholine sensitivity was not possible. DISCUSSION These observations describe a remarkable decrease in cutaneous cholinergic reactivity in subjects sampled

from primitive populations residing rn L~:I-YJlt’fercili environments. The logical reasons for .tAin~~ !!I+ question and its relation tl.1allergic disca:~.. iT1s.b’.s.;i pear obscure. The rationale behind the ..:*::.i!c..’1::. sented m this article stem> t‘rorll i’urlc>biri ;~/>~j~~~ l~l-. sible relationships brtwecn the postulai;c* .~lap!~~.~ function of 1gE.’ ’ the lack ot allergy ii: para\rlc-it.fcsted populations.‘ the clearly maladapr!\;-: ro~b~ iii I&,’ and autonomic dysfunction of pLttL:ni\ v, I:!> ~1 lergic disorders. ’ Srver~l provocati\c .~h,%cr\;ttIi)l!‘* prompted this curiosity. These include ctwct:L rhi ‘,t*runi concentrations of I@ and the severit> OI alicrg~ Ji,ease.-‘ ’ ! 5) the increased prevalence ot’al~t~~ii di:,ca~* in subjects after migration from primitlvc- wrr!wm!ings to urban environments haviny the sxw ilirnart*. (3) the seasonal changes in serum lgt: cortccntration~ detected among patients with allergic iit~~ac.” : G! the preseasonal increase in the ragweed antigen &:x needed to provoke histamine release from ,idopriveI> sensitized basophils.!” and 15) the well d~~cumcnfccl seasonal variation in mechanically induced derma! erythema”‘and plasrna histamine..’ Collec?ivei!. their articles suggest that physiologic differences ~~sociaiied with adaptation to varied ambient tempL-raturci and levels of physical activity niight inlluent.~ pprmcazc release. Our prior studies suggest the iieedom ok the Waorani Indians from allergic disease> and cutaneous rcactions to allergenic pollen extracts is ncjt iau~cd h\ a lack nt’ cutaneous mast ceil5 nor by competitive inhibition between the high circulating con~.cntration~ of polyclonal IgE and pollen-specific 1gF. antibod& for binding sites on cutaneou\ mast cells fhs skin nt’ the Waorani reacts to histamine and can hi: ;~c!optlvet~ sensitized to specifically react to allergenic pollen
VOLUME NUMBER

76 6

Cutaneous ----

and are exposed to seasonal change in a temperate climate. The North Carolina subjects had the greatest prevalence of past allergic and vasomotor symptoms, maximum cutaneous responsiveness to methacholine. and the lowest serum concentrations of IgE. These observations describe variation in two physiologic parameters considered important in the pathogenesis of allergic diseases. Cross-cultural, climatologic. ecologic, and genetic differences among the populations studied could contribute to these interesting differences. At the present level of development of these observations, it is not possible to know how the innate and exposure-related differences between these populations contribute to the observed differences in cholinergic reactivity, nor do we know how these differences relate to the pathogenesis of allergic disease. In spite of this disclaimer, these observations do permit four general conclusions. First, a wide range of variation in cutaneous responsiveness to cholinergic stimuli and circulating IgE concentrations exists among subjects who have adapted to different cultures and environments in various parts of the world. Although it is not possible to know whether cholinergic sensitivity or responsiveness accounts for the absence of reactivity among the Waorani, the lack of significant difference in cholinergic sensitivity between the Tibetan and North Carolina subjects suggests differences in responsiveness are more likely. This interpretation is consistent with several possible mechanisms including ( 1) a decrease in the activity of cutaneous mast cell cholinergic receptors, (3) a decrease in the amount of histamine released by cutaneous mast cells, and (3) an increase in the local catabolism of histamine. These last two possibilities are worth emphasis because of prior evidence of the diminished magnitude of histamine and allergen reactivity in the Waorani,’ the ease of induction and rapid clearance of mechanically induced cutaneous reactive hyperemia during the warm months of the year,‘” and seasonal changes in plasma histamine.” Second, these observations are consistent with the possible existence of an important adaptive relationship between the protective function of IgE antibodies and the cholinergic function of the skin. The basis for this relationship is not clear. Although it would be tempting to postulate that the parasite-induced, hostprotective mechanisms decreased cutaneous cholinergic responsiveness, it is important to note that other equally plausible unresolved alternatives exist. For example, cutaneous cholinergic function is important in the thermoregulatory mechanisms responsible for body heat dissipationZ2 and could influence the responsiveness of skin mast cells to cholinergic stimuli.

methacholine

reactivity

and IgE concentraGons

853

Adaptive physical conditioning to the climate of the rain forest could account for these findings. This possibility is especially attractive in view of the partial independence between atopic disease and parasitic infestation in folk who move from primitive surroundings to a more urban environment.’ Consider alternatively the potential for survival and production of progeny in the tropics of individuals who are susceptible to exercise-induced anaphylaxis.” Selection pressures for inherited mechanisms that permit survival in the tropical rain forest could also account for these interesting findings. Additional observations are needed before these and other interesting possibilities can be resolved. Of most importance, hypotheses focused on these findings and their general implications with respect to the use of favorable climates and conditioning in the care of allergic patients can be subjected to experimental verification. Third, an inverse relationship exists between the reactivity to cholinergic stimuli and circulating IgE concentrations across diverse populations. This interesting relationship could also be detected among the North Carolina subjects, but the magnitude of observed variation at the sample size studied was of marginal statistical significance. The most important implication of this study may stem from the relative clinical value of similar measurements in patients with allergic disease. Measures of cutaneous sensitivity and responsiveness to methacholine and serum IgE and/or IgE antibodies may provide a better predictor of the prevalence and severity of allergic symptoms than measures of either parameter alone. Comparisons between these two parameters in patients and in subjects may provide insight into the interaction between mechanisms that contribute to the severity of allergic disease. Finally, from the perspective of physically active primitive populations who have adapted to warm climates and an appreciable parasite burden, decreased reactivity to cholinergic stimuli and increased circulating concentrations of IgE appear as important as the opposite trends detected in the subjects from North Carolina. These studies would not have been possible without the generousencouragementand support of Dr. D. B. Amos during the predoctoraltraining of Dr. J. W. Larrick and the collaborative help provided by K. Burck. C. Chiles, B. Kane, S. Anderson. K. Dunn, C. Machamer, G. Schlagel and J. Yost. Mr. S. Topgyal provided indispensablehelp during the Tibetan studies. We thank Dr. R. H. Buckley for the measurementsof IgE in the United Statesresidents. and Drs. C. A. Piantadosi and J. A. Kylstra for helpful critical comments about the physiologic implications of the observations. Preparationof the manuscript was made

854

Buckley et al.

possible by the excellent secretarial help of Ms. Joan Murphy, Ms. Dianne Jacobs, Mrs. Pam Rust, and Mrs. Caron Mayhall. REFERENCES I. lshizaka K. lahizaka T: Identification of E antibodies as a carrier of reaginic activity. J Immunol 99: I 1X7. 1967 2. Ltchtenstein LM. Ishizaka K, Norman PS. Sobotka AK, Htll BM: IgE antibody measurements in ragweed hay fever: rclationship to clinical severity and the results of immunotherapy. J Clin Invest 57:473, 197.3 2. Gleich GJ. Jacob BS. Yunginger JW. Henderson LL: Mcawrement of the absolute levels of IgE antibodies in patients with ragweed hay fever. Effect of immunotherapy on seasonal changes and relationship to IgG antibodies. J .AI.LERGY CI.IN lhth~tr~o~.60: 18X. 1977 4. Kaliner M. Shelhamer JH, Davis PB. et al: Autonomic nervous system abnormalities and asthma. Ann Intern Mcd 96:349. 19x2 5. Merrett TG. Merrett J. Cookson JB: Allergy and parasites: the measurement of total and specilic IgE levels in urban and rural communities in Rhodesia. Clin Allergy 6: 13 I. 1976 6. Larrich JW. Buckley CE III. Machamer CE. Schlagel CD. Yost JA. Blessing-Moore J. Levy D: Does hyperimmunoglobulinemia-E protect tropical populations from allergic disease’? J ALLERGY CLIV IMMUNOI. 71:184. 1983 7. Lynch NR. Medouze L, DiPrisco-Fuenmayor MC. Verde 0, Lopez RI. Malave C: Incidence of atopic disease in a tropical environment: partial independence from Intestinal helminthiasia. J ALLERG>’ Cl-IN IhtMUNOI. 73:279, 19X-I 8. Buckley CE 111.Lee KL. Burdick DS: Methacholine-induced cutaneous tlarc response: bivarate analysis of responsiveness and sensitivity. J ALLERGY CL-IN IMMLINOL69:25. 1987 9. Larrick JW. Yost J. Kaplan J. King G. Mayhall J: Patterns of health and disease among the Waorani Indians of eastern Ecuador. Med .Anthrop 3:17-l. 1979 10. Kaplan JE. Larrick JW. Yost J. Calisher CH. Farrell L. Greenberg H. Herrman KL. Sulzev AJ. Walls KW. Pederson L: Infectious disease patterns in the Waorani. an isolated Amcrican Indian population. Am J Trop Med Hyg ?9:298. 1980

17. Buckley CE 111:Atupy and allergy. fr7 Jokltk &‘h. Hillett HP. Amos DR. edihw Zinsscr te.xtbook rrf microbiology. ‘<:2\\ York.. 10X4, Appleton-Centur!-Crufts IX Gleich GJ, %immsrmann EM. Henderson 1.1.. ‘x’rmrgtgrt- JM’. Effect o! immurwthcrapy on immunoglohul~r~ 1: .2nd irnnnnoglohulm G antibodies to ragweed antigen\: 2 ..i>-yc;tr pi<, spective study J .4.LI:RG’l CL11 iMZIL~NOI ‘70:X!. I%2 19. Petry R. Swcnwn R. Busx W. Seasonal vartatton rn human hasophilic leukocyte histamine r&ease to rag~w:i :rntigen F, (.$$I.

.I Al.1 iiR(;\r

ci

IN IMhll’KC)L

:i:I%t,

!%d

70. DiPalma JR, Reynolds SRM. I~o~tcr FI: t&xttttritt\c mea wrement ot reacttvc hyperemia in human \kir; -Irn Hear1 i 2.::377. 19-12 71 Hasegavva M. Saito \ Naka f.. Ktyrir M: Seawndi vanawns of total histamine in patient\ with wasonal allerstc rhinitt\ Chn Allergy 13:177, 19X3 21. Fox RH. Edholnt OG: Nervous control of cutaxous Lircuiation. Br Med Bull 191 IO. 1963 23. Kaplan 4P. Natbony SF. Tavr 11 Al’. Fruchter iL. Foster bl. Eserws-induced anaphylaxts as a manifestation :r! iholincrgt~ urticarta I .Al.l.t.~(;>~Cl IN IRfMIIv(aL 68:319. IGJR!