Evaluation of the cause of nasal and ocular symptoms associated with lawn mowing

Graver

J. ALLERGY CLIN. IMMUNOL. MAY 1986

et al.

29. Lowry OH, Rosenbrough NJ, Farr AL, Randall RJ: Protein measurement with the folin phenol reagent. J Biol Chem 193:265, 1951 30. Weksler BB, Pett SB, Alonso D, Richter RC, Stelzer P, Subramanian V, Tack-Goldman K, Gay WA Jr: Differential inhibition by aspirin of vascular and platelet prostaglandin synthesis in atherosclerotic patients. N Engl J Med 308:800, 1983 31. Brater DC, Peters MN, Eshelman FN, Richardson CT: Clinical comparison of cimetidine and ranitidine. Clin Pharmacol Ther 32:484, 1982 32. Guo Z-G, Levi R, Graver LM, Robertson DA, Gay WA Jr: Inotropic effects of histamine in human myocardium: differentiation between positive and negative components. J Cardiovasc Pharmacol 6:1210, 1984 33. Ishizaka K, Ishizaka T: Immunology of IgE-mediated hypersensitivity. In Middleton E Jr, Reed CE, Ellis EF, editors: Allergy: principles and practice. St. Louis, 1983, The CV Mosby Co, p 43 34. Mehta J, Mehta P: Prostacyclin and thromboxane A, production by human cardiac atrial tissues. Am Heart J 109: I, 1985

Evaluation symptoms Michael Detroit,

S. Rowe,

35. Roberts LJ II, Sweetman BJ, Lewis RA, Austen KF, Oates JA: Increased production of prostaglandin D, in patients with systemic mastocytosis. N Engl J Med 303: 1400, 1980 36. Roberts LJ II, Lewis RA, Oates JA, Austen KF: Prostaglandin, thromboxane, and l2-hydroxy-5,8,10,14-eicosatetraenoic acid production by ionophore-stimulated rat serosal mast cells. Biochim Biophys Acta 575: 185, 1979 37. Levi R, Malm JR, Bowman FO, Rosen MR: The arrhythmogenic actions of histamine on human atrial fibers. Circ Res 49:545, 1981 38. Wolff AA, Levi R: Histamine and cardiac arrhythmias. Circ Res 58:1, 1986 39. Stein I, Wecksell I: Cardiac disease accompanying allergic drug reactions. J ALLERGY 45:48, 1970 40. Austin SM, Barooah B, Kim CS: Reversible acute cardiac injury during cefoxitin-induced anaphylaxis in a patient with normal coronary arteries. Am J Med 77:729, 1984

of the cause of nasal and ocular associated with lawn mowing M.D., Judith

Bailey, B.S., and Dennis R. Ownby,

M.D.

Mich.

A number of individuals with perennial or seasonal rhinoconjunctivitis state that their symptoms may suddenly worsen on exposure to lawn mowing. Many allergists have believed that this was related to the agitation of molds deposited on the grass. We studied 50 consecutive new patients with rhinitis using history, skin testing, total and specific IgE assays, and nasal smears. Twenty-jive patients gave histories of minimal or no change in their nasal symptoms with exposure to lawn mowing, whereas 11 patients had dejinite, but mild, symptoms, and 14 patients had severe symptoms. Positive skin tests to grasses, trees, and weed pollens were more frequent in those patients with symptoms exacerbated by lawn mowing (p < 0.03). Symptomatic patients also had higher serum concentrations of total IgE (p < 0.008) and grass-pollen specljic IgE (p = 0.0004). The frequencies of positive skin tests to ragweed pollens, house dust, molds, and grass leaves, as well as the percentage of nasal eosinophils, were not different in the symptomatic and asymptomatic groups. No signiJcant association was found between symptoms and IgE antibodies to molds or grass-leaf extract. We believe that the increased nasal and ocular symptoms coincident with lawn mowing are allergic phenomena significantly associated with skin test sensitivity and specific IgE antibodies to grass pollens but not with sensitivity or spec$cIgE to molds or grass-leaf extract. (JALLERGY CLINIMMUNOL 77:714-19,1986.)

From the Division of Allergy and Clinical Immunology, Departments of Medicine and Pediatrics, Henry Ford Hospital, Detroit, Mich. Supported by the Fund for Henry Ford Hospital. Received for publication Feb. 14, 1985. Accepted for publication Oct. 24, 1985. Reprint requests: Michael S. Rowe, M.D., Division of Allergy and Clinical Immunology, Henry Ford Hospital, 2799 West Grand Blvd., Detroit, MI 48202. 714

Many individuals with symptoms of rhinitis or conjunctivitis complain that their symptoms are exacerbated when they are exposed to lawn mowing. The degree of sensitivity ranges from those individuals who state they sneeze occasionally when they mow their lawns to individuals who state that they develop severe symptoms if a neighbor mows his lawn. The cause of the increased symptoms related to lawn mowing has not been defined, but some authorities have

VOLUME 77 NUMBER 5

Cause

/Microtiter

of symptoms

with

lawn

mowing

715

well surface BtOTlNYLATED [a] anti - IgE +

Read AP Fluorscence 4 assay

b

FIG. 1. Diagrammatic representation irrlmunosorbent assays system used

of the avidin-biotin in this study.

pffF1

PBS: Phosphate-buffered salute

suggested that symptoms are related to mold exposure. ’ Presumably, molds grow in the detritus on the ground under the grass leaves and are aerosolized by the air currents created by power lawn mowers. The purpose of this study was to determine whether symptoms occurring in association with lawn mowing

could be related to the results of allergen skin tests, nasal smears, and levels of total and allergen-specific IgE. We found that increased symptoms with lawn mowing were highly associated with sensitivity to grass pollens, but not with sensitivity to common

molds.

amplified

fluorescent

enzyme-linked

yet those patients classified as having minimal symptoms may be considered relatively asymptomatic, since they did not have a noticeable increasein rhinitis symptomatology with exposure to lawn mowing. Those patients in the mild and severe groups may be considered symptomatic from exposure to lawn mowing becausethey noticed a definite increasein their chronic symptoms. Each patient also had skin tests performed for common inhalant antigens, including four grasses(Kentucky bluegrass, orchard, redtop, and timothy) and IO molds (Alternaria, Helminthosporium, Cladosporium, Aspergillus, Penicillium, Curvularia, Fusarium, Pullularia, Mucor, and Rhizopus) (Greer Laboratories, Lenoir, N. C.). Alternaria, Cladosporium, Helminthosporium, and Fusarium are the

most common field fungi.‘. ’ Skin tests by use of either the prick test (1: 20 w/v) or intradermal (1: 1000 w/v) methods were read aspositive if 2 + (>2 I mm erythema) or greater.4 Intradermal testing was performed with mold groupings, whereasprick testing used single antigens.

IgE assays MATERIAL Subjects

AND METHODS

Fifty consecutivenew patients referred for evaluation of rhinitis were examined during the summer months of 1983. Young children who could not relate concisehistories were excluded, as well as patients who did not have symptoms of rhinitis or conjunctivitis. Patients had standard history and physical examination and were specifically questioned regarding any change in nasal or ocular symptoms when they were exposed to freshly mowed lawns. Symptoms associatedwith mowing were graded asminimal, if the patient noted little symptomatic change with exposure; mild, if the patient noted ;i definite increase in sneezing, tearing, or itching with exposure but the symptoms were not severe enough to prevent the patient from either mowing the lawn or being near lawn mowing; and severe,if the patient stated that they could not be near lawn mowing becauseof markedly increasedsymptoms that would last for more than 1 hour after exposureor definite symptoms after minimal exposure. Histories were recorded before skin testing. It should be noted that all patients were symptomatic since they were referred for the evaluation of chronic rhinitis, and

Total and allergen-specificserum IgE levels were measured by use of an enzyme-linked immunoassay that is illustratedin Fig. 1.5For allergen-specificIgE measurements, 12-well polystyrene microtiter strips (Micro-Fluor “B” removawell, Dynateck, Inc., Alexandria, Va.) were coated with extracts of each allergen. The allergens studied were timothy, Kentucky bluegrass,Alternaria, and Cladosporium (Hollister Steir, Downers Grove, Ill.). Timothy and Kentucky bluegrasswere chosen as representativesof the common northern grasses.For representativemolds we chose Alternaria becauseit is the outdoor mold most commonly eliciting positive skin test reactions and Cladosporium becauseit is the most prevalent mold in outdoor air samples.‘. ’ For coating, the extracts were deglycerinized by passageover a PD-10 column (PharmaciaInc., Piscatawa, N.J.) and then diluted in a 60 mmol/L carbonate-bicarbonate buffer, pH 9.8. One hundred microliters of previously determined optimal dilutions were incubated in each well for 18 hours at 4” C. With pools of positive sera, the optimal dilution for the grass extracts was 1: 100 and 1: 50 for the mold extracts. The wells were then washed with PBS and blocked with FCS-PBSfor 1 hour at 4” C to prevent non-

716

Rowe

J. ALLERGY CLIN. IMMUNOL. MAY 1986

et al.

4 a c z *: ‘; 7 0 6L g 2

100 90 80

E

10

0 &

25

r P 20 .$ :: 15 t J 0 10 ct b 5 5

70 60 50 40 30 20

E s Tree

Grass Pollen

Pollen

History History FIG. 2. Relationships groups of pollens

TABLE

I. Patient

RzKd

PFICn

positive negative

for grass for grass

of the percentage and the presence

symptoms= symptoms

Mold

“D”““ue

l pe

0.03

0

and number of patients or absence of symptoms

with positive associated

skin tests with lawn

to different mowing.

characteristics Lawn Minimal

No. of patients M/F Mean age (SD) Geometric mean Total serum IgE* Mean nasal eosinophils No. with nasal eosinophils 2 4% *p < 0.03 for differences

between

the three

groups

Severe

11 318 24 (10.8) 118

5% 6

14 915 23 (9.6) 164

7% 3

of variance);

specific absorption of other proteins. The wells are then stored with 200 cl.1of FCS-PBScontaining 0.1% Na azide. The wells were washed for use once with FCS-PBS, and 100 pl of patient sera was added to triplicate wells and incubated over night. The wells were then washed three times with FCS-PBS, and 100 ng of biotinated affinitypurified antihuman IgE in 100 pl of 1% FCS-PBSwasadded to the wells for 1 hour at room temperature. The affinitypurified anti-IgE (Kirkegaard-PerryLaboratories, Gaithersberg, Md.) was biotinated by reacting with biotinyl-hr-hydroxysuccinimideat 1: 1 molar ratio asdescribedby Kendal et al.’ The anti-IgE that does not bind was washedfrom the wells, and 100 p,l of a 1: 5000 dilution of avidin-coupled alkaline phosphatase(SigmaChemical Co., St. Louis, MO.) was added for 20 minutes. The wells were again washed three times with PBS-FCS,and 100 ~1 of diethanolamine buffer, pH 9.8, was added for 10 minutes followed by another washing of the wells. The enzyme substrate, 100 pl of 10m4mol/L 4-methylumbelliferyl phosphate, was then added, and the resulting fluorescencewas determined by reading the wells in a model 100 microfluorimeter (Allergenetics Inc., Mountain View, Calif.) at intervals from 10

symptoms

Mild

25 14/11 19 (12.9) 56

(analysis

mowing

for minimal

versus

mild

9% 4 and severe,

p = 0.008

(t test).

to 60 minutes. The readingsare expressedasmultiples above control wells to which sera from nonallergic adults have been added. Total serum IgE was determined in a manner analogous to that for specific IgE except that the wells are initially coated with the antihuman IgE antibody. Standard curves were constructed with appropriate dilutions of the U. S. reference for human IgE’ and dilutions of a patient serum standardizedagainst the U. S. reference standard. The averagenet fluorescentunits producedby appropriatedilutions (typically 1: 25 and 1: 250) of the patient’s sera were converted to units of IgE by comparison to the standard curve. The resultsof total IgE assaysin the seraof 60 other patients were compared in this ELISA and in the PRIST assay(Pharmacia Diagnostics), and the correlation was 0.96. Grass leaf-extract

preparation

Fifty grams of grass leaves were obtained with a hand clipper from a lawn of Kentucky bluegrass.The leaveswere washed lightly in water and were then mixed in a blender with 1000 ml of PBS for 5 minutes. The PBS and grass leaves were then slowly mixed for 10 hours at 4” C. The

VOLUME 77 NUMBER 5

TABLE and --

Cause

II. Relationship skin

tests

to

of grass grass

pollen

symptoms and

of symptoms

with

MILD

MINIMAL

molds

mowing

717

SEVERE

150 r

Grass Minimal

i

symptoms Mild

Severe

= 2; X’ = = 2; X2 =

I5 IO 5

I IO ii

I I3 iT=

xl

5

4

17 33 iii 21

15.07;~ = 0.0005. I .46; p = 0.48.

pulp was removed by filtration through cheesecloth,and the extract was cleared by centrifugation at 6000 rpm for 30 minutes at 4” C. The I :20 w/v extract was mixed with an equal volume of glycerol, producing a fluid extract of I :40 w/v in 50% glycerol. This extract was sterilized by passagethrough a 0.45 micron filter and was used for skin testing. Aliquots of the extract without glycerol were used at 1: 100 concentrationsto coat wells for assayand at I : 20 in attempts to inhibit binding of patient IgE to timothy grass pollen-coated wells.

Statistical

methods

Statistical analysisfor comparison of total and allergenspecific IgE levels between patient groups were performed by use of logarithmically transformed data. Both Student’s t test and analysis of variance were used. Data from contingency tables were analyzed by use of the chi-squarestatistic.

RESULTS The characteristics of the patients with and without grass symptoms are illustrated in Table I. The pro-

portions of male and female patients in the minimal, mild, and severe groups were not significantly different (p = 0.16 by chi-square). Mean age was slightly higher in the symptomatic (mild plus severe) group. Total serum IgE was significantly higher in the symptomatic group (p = 0.008). Each groupdemonstrated a similarly sized minority of subjects with nasal eosinophilia 24%. As can be observed in Fig. 2, the group with symptoms on exposure to mowing had significantly higher frequencies of positive skin tests to grasses (p = O.OOOS),trees (p = O.OOl), and weeds (p = 0.024) compared to the asymptomatic group. This was not observed for molds (p = 0.48), ragweeds (p = 0.11 I), or house dust (p = 0.24). Of these 50 pa-

8

;

0

. -

.

:

I2

i

Total t

Grass-pollenskin test* Negative Positive Total Any mold skin test? Negative Positive Total *df ‘rdf

lawn

-

-

0 0

T-

T -:

*

.

-

s 0

. 8 .

0

80

. 0

. .

O

O

.

FIG. 3. Relative quantities of grass pollen-specific IgE in serum samples of patients grouped by class of symptoms. The so/id dots represent timothy grass, and the circles represent Kentucky bluegrass-specific IgE. The bars indicate the geometric means and 1 SE about the means.

tients only five had symptoms of asthma or exerciseinduced bronchospasm, and these symptoms were not related to the presence or absence of symptoms from lawn mowing. None of the patients had symptoms of other allergic diseases such as eczema or urticaria. Table II compares the number of patients with positive skin tests to grass pollens or molds with the class of grass-exposure symptoms. Only two of 17 patients with negative grass pollen skin tests complained of grass-exposure symptoms, whereas 23 of 25 patients who complained of grass symptoms had positive skin tests to grass pollens. These proportions are highly significant (p = 0.0005). In contrast, only 16 of 25 patients who complained of grass symptoms had a positive mold skin test. Nine of 21 patients with a negative mold skin test had grass symptoms. These proportions are not significant (p = 0.48). Fig. 3 illustrates the levels of grass pollen-specific IgE in three groups of patients separated by the degree of symptoms. The specific IgE levels for timothy and Kentucky bluegrass pollens were significantly correlated (r = 0.89, p < 0.00005). The 25 patients with

718

J. ALLERGY CLIN. IMMUNOL. MAY 1986

Rowe et al.

5

MINIMAL

TABLE III. Skin testing and specific IgE to grass pollen and grass leaf in patients with severe symptoms Present

Specific IgE (n = 10 patients) (in vitro) Grasspollen Grassleaf Positive skin tests (n = 12 patients) Grasspollen Grassleaf Mold

Absent

10 0

0 10

12 0 5

0 12 I

tests to the mold, but none of the 12 had positive skin tests to the grass leaf. FIG. 4. Relative quantities of mold-specific IgE in serum samples of patients grouped by class of symptoms. The solid dots represent Alternaria. and the open circles represent Cladosporium-specific IgE. The bars indicate the geometric means and 1 SE about the means. The mean levels among the groups do not differ significantly.

minimal symptoms had significantly lower levels of grass pollen-specific IgE than the 25 patients with mild or severe symptoms (p = 0.0002 and p = 0.008 for timothy and Kentucky bluegrass, respectively). Because of limitations of the original serum samples, assays for Kentucky bluegrass IgE were possible for only 47 patients. Fig. 4 presents the results of specific IgE measurements to the two mold extracts. The difference between the three patient groups and between the patients with minimal compared to those with mild and severe symptoms are not significant (p > 0.50 for all comparisons with 49 assays for Alternaria and 48 assays for Cladosporium). After finding the significant associations between symptoms on exposure to grass mowing and grasspollen sensitivity, we believed that grass allergen might be contained in the grass leaves. Ten patients with severe symptoms were examined for specific IgE to grass pollen and grass leaf (Table III). All demonstrated positive ELISA values for grass pollen, but no significant reactivity was detected to the grass-leaf extract. We also tried to inhibit the response to grass pollen by adding 50 ~1 of grass-leaf extract to each 100 pl of patient serum. The addition of the grassleaf extract failed to produce inhibition. To investigate further this question, we selected 12 patients with histories of severe symptoms and prick skin tested them to timothy grass pollen, Kentucky bluegrass leaf, and Alternaria extract. All 12 patients had positive skin tests to the grass pollen, five had positive skin

DISCUSSION The purpose of this study was to determine the probable cause of the sudden increases in nasal and ocular symptoms when patients are exposed to freshly mowed grass. Our findings suggest an allergic etiology since the symptomatic group had significantly higher levels of total IgE and higher percentages of positive skin tests to grass, tree, and weed pollens. Grass pollen-specific IgE was also significantly higher in the symptomatic group. Mold sensitivity has been believed by some allergists to be the cause of symptoms associated with grass mowing, perhaps by aerosolization of dormant molds when mowing occurs. This does not appear to be the most likely cause, since there were no significant differences in the frequency of positive skin tests for mold or the quantities of mold-specific IgE. In contrast to reports of allergen in plant parts other than pollen, which has been reported with ragweed,’ we were not able to document a positive skin test or specific IgE to the grass-leaf extract in symptomatic subjects. This was true even in patients with severe symptoms when they were exposed to lawn mowing and who were highly sensitive to grass pollen. The association between symptoms around grass mowing and the sensitivity to grass pollen suggests that these symptoms- are due to grass-pollen allergy even though grass that is mowed regularly usually does not pollinate. One possible mechanism for this association is that grass-pollen grains settle out of the air onto lawns and are then propelled back into the air by lawn mowers.3 Such a mechanism would also explain why most patients continued to have symptoms of similar severity around lawn mowing outside of the grass-pollen season, since these symptomatic

VOLUME 77 NUMBER 5

patients were significantly more likely to have positive skin tests to other pollens. These possibilities will need to be examined by air sampling around lawn mowing. Nonallergic mechanisms could also be responsible for the increased symptoms. Allergic individuals may be more susceptible to nonspecific irritation produced by soil particles and plant debris becoming airborne as a result of lawn mowing. This mechanism, however, would not explain why other highly allergic individuals, such as those sensitive to ragweed, do not have symptoms. In conclusion, it is probable that increases in nasal and ocular symptoms associated with lawn mowing are of allergic origin and related to sensitivity to grass and other pollens. We gratefully acknowledge the secretarialassistanceof Ms. June Johnston. REFERENCES I. Solomon WR, Matthews KP: Aerobiology and inhalant allergens. In Middleton E, Reed CE, Ellis EF, editors: Allergy:

Cause

2.

3.

4.

5.

6.

7. 8.

of symptoms

with

lawn

mowing

719

principles and practice, ed 2. St. Louis, 1983, The CV Mosby Co, p 1180 Gutman AA: Allergens and other factors important in atopic disease. In Patterson R, editor: Allergic diseases, diagnosis and management. Philadelphia, 1980, JB Lippincott Co, p 135 Solomon WR, Matthews KP: Aerobiology and inhalant allergens. In Middleton E, Reed CE, Ellis EF, editors: Allergy: principles and practice, ed 2. St. Louis, 1983, The CV Mosby Co, pp 1181-2 Booth BH III: Diagnosis of immediate hypersensitivity. In Patterson R, editor: Allergic diseases, diagnosis and management. Philadelphia, 1980, JB Lippincott Co, p 88 Rao PVS, McCartney-Frances NL, Metcalfe DD: An avidinbiotin ELISA for rapid measurement of total and allergen-specific human IgE. J Immunol Methods 57:7 I, 1983 Kendall C, Ionescu-Martiu I, Dreesman GR: Utilization of biotin/avidin system to amplify the sensitivity of the enzyme-linked immunosorbent assay (ELISA). J Immunol Methods .56:329, 1983 Evans R III: A U. S. Reference for human immunoglobulin E. J ALLERGY CLIN IMMUNOL 68:79, 198 I Agarwal MK, Swanson MC, Reed CE, Yunginger JW: Airborne ragweed allergens: association with various particle sizes and short ragweed plant parts. J ALLERGY CLIN IMMUNOL 74:687, 1984