Variability in parameters of allergic bronchopulmonary aspergillosis in patients with cystic fibrosis

Variability in parameters of allergic bronchopulmonary aspergillosis in patients with cystic fibrosis Patricia S. Hutcheson, BA, Anthony J. Rejent, MD, and Raymond G. Slavin, MD St. Louis, MO. Seventy-nine patients with cysticjbrosis (CF) were evaluated and were followed in a longitudinal, prospective fashion during a 6-year period for the development of immune parameters indicating Aspergillus fumigatus (Af ) sensitization and allergic bronchopulmonary aspergillosis (ABPA). Although four patients developed frank ABPA, there was considerable variability in immune parameters in non-ABPA. Twenty-four patients became skin test positive to Af with none losing skin reactivity. Twenty-ftve patients developed serum precipitins to Af, whereas 12 patients lost their precipitins. Of 15 patients with an elevated total serum IgE of 22 SD, five demonstrated a marked decline of at least 40%. Three of 16 patients with IgE-Af became negative, whereas eight of 27 patients lost their IgG-Af. None of these patients had received corticosteroid therapy that could have accounted for the findings. Thus, patients with CF frequently lose evidence of Af sensitivity spontaneously without corticosteroid intervention. The diagnosis of ABPA in CF should not be based solely on serology and skin test results, since at any point in time, patients with CF may demonstrate variable responses to Af. (.I ALLERGY CLIN IMMUNOL 1991;88:390-4 .) Key words: ABPA, Aspergillus antibodies, cystic fibrosis

ABPA was first described in patients with asthma in 1952.’ More recently, it has also been reported to occur in patients with CF with an incidence of 10% to 12%.‘” Early diagnosis of ABPA with prompt administration of systemiccorticosteroids is essentialto preventthe developmentof bronchiectasis,pulmonary fibrosis, and pulmonary insufficiency.’ Becauseof the clinical and laboratory characteristicssharedby ABPA and CF (Table I), the diagnosisof ABPA complicating CF is often difficult to make. We instituted a prospective longitudinal study of patients with CF to determine if any in vivo or in vitro pattern might emerge that would enable us to make an earlier diagnosis of ABPA. Despite the fact that four patients of the 79 studied have developed ABPA, no clear pattern has

Abbreviations used

ABPA: Allergic bronchopulmonary aspergillosis CF: Cystic fibrosis Af:

IgE-Af: IgG-Af: BSA: ST: ppn: Ag: OD: Ab:

Aspergillus fumigatus IgE antidspergillus fumigatus IgG antidspergillus fumigatus

Bovine serum albumin Skin test Precipitin Antigen Optical density Antibody

emerged;in fact, we have noted a surprising variety of responses. From the Division of Allergy and Immunology, Departments of Internal Medicine and Pediatrics, St. Louis University School of Medicine, St Louis, MO. Supported by National Institutes of Health Grant HL 30652. Received for publication Nov. 12, 1990. Revised April 19, 1991. Accepted for publication April 21, 1991. Reprint requests: Patricia S. Hutcheson, St. Louis University School of Medicine, 1402 S. Grand Blvd., Room R209, St. Louis, MO 63104-1028. 111130593

390

MATERIAL AND METHODS Patient population The patient population consistedof 79 patients observed in the CF clinic at Cardinal Glennon Children’s Hospital, St. Louis, MO. The diagnosis of CF was confirmed with an iontophoric sweat chloride level of 260 mEq/L on a pilocarpine-induced sweat sample of X0 mg.8 Institutional Review Board approval was granted for the study, and informed consentwas obtainedfor eachpatient. Patients’ ages

VOLUME NUMBER

ABPA parameters

88 3, PART 1

TABLE I. Clinical and laboratory

TABLE II. Variability

characteristics

in patients

shared

by ABPA and CF

ranged from 2 to 30 years, and the sex ratio was 43 boys/ 36 girls. The 79 patientswere all evaluatedmore than one time; most were studied at yearly intervals, but in some instances, studies were done either more or less often. The diagnosis of ABPA was made according to the criteria set forth by Rosenberg et a1.9as modified by Laufer et al5 These included episodic bronchial obstruction (asthma),peripheral blood eosinophilia, positive immediate epicutaneous ST to Af, positive serumppn to Af, a serumtotal IgE level of 2 1000 IU/ml, elevated Af-specific IgE and IgG serumlevels, a history of pulmonary infiltrates, and central bronchiectasis. The presenceof the first sevencriteria was required to makethe diagnosisof ABPA. The diagnosiswas made in four of the 79 patients. Parameters In vivo immediate epicutaneous skin testing was performed with Af (1:20, Greer Laboratories, Inc., Lenoir, N.C.) and other St. Louis area inhalants, including trees, grasses,ragweed,Dermatophagoidesfarinae, dog, cat, and Candida. Appropriate histamine and saline controls were also performed. Results were graded 1 to 4 + with positive designatedat 22 + . In vitro total IgE (international units per milliliter) was determinedby PRIST (PharmaciaDiagnostics, Piscataway, N.J.); ppns were demonstratedwith Ouchterlony templates (Greer Laboratories, Inc.) and Ags that included various Af preparations (D&it, Greer Laboratories, Inc., and private source) and Candida albicans. Patient serum (undiluted) was placed in the center well and Ags in the outer wells; ppns were read at 48 hours and were consideredpositive if a ppn line developed between the Ab and Ag wells of at leastoneAf preparation.Rabbit anti-Af (GreerLaboratories, Inc., Diakit) was included as a positive control. Specific IgE anti-Af (IgE-Af) and IgG anti-Af (IgG-Afl were performed as described according to the method of Greenbergerand Patterson”’ with some modification. The IgG-Af assayconsistedof placing 2.5 pg/ml of Ag (Greer Laboratories, Inc., lyophilized lots LM3-16 and LM3-26) in carbonatebuffer (pH 9.4) into flat-bottomed, microtiterplate wells overnight at 4” C. The plates were washedwith BSA-saline (1% BSA and 0.05% Tween in 0.85% saline). Dilutions of serum (BSA-saline) were added, incubated 2 hours at 37” C, and washed. Goat antihuman IgG alkalinephosphataseconjugate (I: 1000, Sigma Chemical Co., St. Louis, MO.) was added; the plates were incubated and

391

of parameters observed with CF and without ABPA

A~OPY

Asthma Positive ST to Af Positive ppns to Af Elevated total IgE Pulmonary infiltrates Bronchiectasis

in CF

Parameter* Positive Af ST Positive Af ppns Positive IgE-Af Positive IgG-Af Total IgE 22 SD Total patients Parameter? Loss of Af ST Loss of Af ppns Loss of IgE-Af Loss of IgG-Af Decline in total IgE Total patients

No.

(%I

24 25 16 27 15 75

(32) (33) (21) (36) (20)

0

12 3 8 5 20

(0)

(48) (19) (30) (33)

*Parametersfound to be positive at any time during follow-up. tparameters found to be negative subsequentto being positive on a previous evaluation.

washed. Phosphatasesubstrate(1 mglrnl in 10% diethanolamine buffer and 0.1% 0.05 mol/L of MgCI,) was added. Plates were incubated 30 minutes at room temperaturebefore stopping the reaction with 5 N NaOH. OD was read at 405 nm on a Bio-Tek microplate reader(Bio-Tek Instruments, Inc., Winooski, Vt.). The IgE-Af assay differed slightly. The Ag concentration was 100 kg/ml, and incubations were performedfor 90 minutes at room temperature. Rabbit antihuman IgE (1: 1000, Calbiochem-Behring, La Jolla, Calif.) was added rather than antihuman IgG, followed by the addition of sheepantirabbit IgG alkaline-phosphataseconjugate(1: 1000, SigmaChemicalCo.). Substrate was added,and the remainderof the assaywas asdescribed. IgG-Af serum dilutions were 1:500 and 1: 1000; IgE-Af serum dilutions were 1: 50 and 1: 100. Sampleswere rnn in triplicate and were considered positive if the ratio of patient OD was at least twice the OD of an ST negative, precipitin negative, control pool used throughout the study. RESULTS

The variable results observed in 75 patients without ABPA/CF during the 6-year period are depicted in Table II. These patients have been evaluated at least twice and are consideredto be non-ABPA, either becauseimmune parametersfor Af sensitization are absent or changes in pulmonary status have been responsiveto antibiotic or increasedbronchodilator therapy. Twenty-four patients (32%) have become ST positive, with none losing their ST sensitivity. Twenty-five patients (33%) developedppns to at least one Af Ag preparation, whereas 12 patients (48%) exhibited a loss of ppns. Sixteen patients (21%) had positive IgE-Af, but three patients (19%) demon-

392 Hutcheson et al.

J. ALLERGY

developed positive ppns that reverted to negative 1 year later. He had positive STs to tree and ragweed but was negative to grass, D. farinae, and

TABLE III. Variability in two representative patients Total IgE (NJ/ml)

Af ST

IgE-Af

IgG-Af

Ppn

20 21 21 22

380 880 920 1850 665 495

+ + + + + +

-

+ + -

-

Patient2t 14 15 16 17

243 800 480 540

+ + + +

+ -

+ -

+ -

Age (yr)

Patientl* 17 19

CLIN. IMMUNOL. SEPTEMBER 1991

*Other skin test results: negative to tree, grass, mite, ragweed, dog, and Candida. TOther skin test results: negative to grass, mite, and Candida and positive to tree and ragweed.

strated a loss. Twenty-sevenpatients (36%) had positive IgG-Af with eight patients (30%) losing this parameter. Thirty-two patients had peak total serum IgE rl SD from the mean of expected values according to age, with 15 patients having ~2 SD.” Serum IgE spontaneously declined in five of these 15 patients from 40% to 93% (X = 52%). The peak IgE values in these patients ranged from 460 to 2600 IU/ml (Z = 1316), and the time period during which this decline occurred was 1 to 6 years (X = 2.4). Four of the five patients were older than 10 years at the time of their peak IgE value. None of these patients had begun corticosteroid therapy that could have accounted for the decline. Parametersfrom two representativepatientsarepresented in Table III. Patient 1 was studied six times between the ages of 17 and 22 years. He was Af ST positive, his total IgE rose to 1850IU/ml, and he had developedIgG-Af Abs by age 21 years. At that time he had negative Af ppns, negative IgE-Af, and was ST negative to tree, grass,D. farinae, ragweed, dog, and Candida. Four months later, his Af ST remained positive, but his total IgE had fallen to 665 IU/ml, and his IgG-Af had becomenegative. One year later, his IgE had fallen to 495 IU/ml. Patient 2 had developed a positive Af ST, an IgE of 800 IU/ml, positive IgE-Af, and positive IgG-Af by 15 years of age. However, 1 year later, his IgE had declined to 480 IU/ml, and his positive IgG-Af and IgE-Af had reverted to negative. At that time he

Candida.

In Tables IV and V, the parametersare compared of four patients with CF in whom the diagnosis of ABPA has been establishedsince the study beganand five patients with CF whoseparametersare suggestive of ABPA. In the patient group before developing ABPA (Table IV), none had an IgE >lOOO W/ml or positive IgE-Af before developing acute symptomsof increasedcough or wheezing, decreasedFEV, , or increasedinfiltrates on chest x-ray film. The increased pulmonary symptoms were unresponsive to conventional therapy, and these patients developedpositive parametersin all five categories. The time between samples at “pre-ABPA diagnosis” and “ABPA diagnosis” varied; it was 14 months for patient 1, 18 months for patient 2, 4 months for patient 3, and 23 months for patient 4. In the group of five patients whose parametersare suggestiveof ABPA (Table V) all have positive STs, ppns, IgE-Af, and I@-Af. Two patients have an IgE > 1000 IU/ml. Two of these patients experiencedan increasein pulmonary symptoms but respondedto conventional therapy. (Patient 3 is a noncompliant patient, and his pulmonary status improved during hospitalization.) No patient hasbeen started on corticosteroid treatment. DISCUSSION This study was undertaken in an attempt to prospectively identify patients with CF at risk of developing ABPA by following five immune parameters indicative of Afsensitization in a large population with CF. The published incidence of 10% to 12% ABPA in CF is dependenton the definition of ABPA, and since the criteria used for the diagnosis of ABPA can often be observedin uncomplicatedCF, the diagnosis of ABPA in CF may be quite difficult. During this 6-year prospective study, four patients with CF did develop ABPA. In addition, a number of patientsdevelopedan increasein various immunologic parametersindicating sensitizationto Af, including ST reactivity to Af, serumppns to Af, an increasein total IgE, and increasesin IgE and IgG anti-Af Ab. Surprisingly, 20 of the 75 patients without ABPA demonstrated a spontaneous decrease in one or more of the above-mentionedparameters,as presented in Table II. By the term “spontaneous,” we mean that none of these patients was receiving corticosteroids. It is well-known that corticosteroid therapy will result in a decreasein total IgE and possible disappearanceof serum ppns.12Particularly notable

VOLUME NUMBER

ABPA parameters in CF 393

88 3. PART 1

TABLE IV. Parameters Pre-ABPA

of patients

with CF before and after ABPA diagnosis

diagnosis

ABPA diagnosis

Parameters

Symptoms*

$ 8‘2

3 Pt

ST

1++ 2++ 3 4++

Ppn

-

+

IgE-Af

IgG-Af

IgE

_ -

+ + + +

140 700 620 555

Parameters

0”

3 *E

&?

-

t t

t t t

-

5e

3

1 4 1 J

t t t t

ST

Pm

IgE-Af

IgG-Af

bE

+ + + +

+ + + +

+ + + +

+ __ -t-

4200 3800 1600 1000

*Unresponsive to conventional therapy.

TABLE V. Five individual

patients

with CF whose

parameters

are suggestive

of ABPA

Parameters

1 2 3 4 5

+ + + + +

+ + + + +

+ + + + +

Symptoms*

+ + + + +

170 410 350 1200 2900

-

-

y

I

t

-

-

ND ; -

ND i -

ND, Not done. *Responsive to conventional therapy.

was a spontaneousdecline in total serum IgE in five patients that was disproportionate to that which could be expected with increasing age. The most dramatic occurred in a ‘I-year-old patient whose peak IgE of 2600 IU/ml dropped to 170 IU /ml during a 6-year period of time, a fall of 93%. Little is known of the changesin immunologic parametersthat may occur in time in patients with untreated ABPA. Safirstein et al.” preformed a retrospective study of 50 patients with ABPA during a 5year period and demonstratedthat untreated patients had a chronic course. This study, however, only tracked clinical symptomsand signs and not serologic measurements.No other study of ABPA in CF, previously noted,2-7has examined changes in immune parametersthat may occur with time spontaneously, i.e., without corticosteroids. Our study indicates that at any point in time, patients with CF may demonstratevariable responsesto

Af and may experiencea diminution in immune reac-

tivity to Afwithout steroid intervention. The diagnosis of ABPA should not be made on the basis of ST and serology alone. Worsening of the respiratory state in a patient with CF with positive parametersof Af sensitization does not necessarily mean that ABPA has developed.As observedin our study, supportive measures, including antibiotics and bronchodilators, will often result in an improvement in the respiratory condition with a decreaseover time in the immune parametersindicative of Af sensitization. In light of our findings, we would hesitate to diagnose ABPA in patients who only possessthe serologic characteristicsof the disease.The absenceof sustained,unresponsive,pulmonary deterioration and the spontaneousreversalof many of the serologiccharacteristics of ABPA suggeststo us that systemic corticosteroids are not necessarily indicated in these patients.

394 Hutcheson et al.

We thank Maria J. Weingartner for her excellent secretarial expertise.

REFERENCES 1. Hinson KFW, Moon JA, Plummer NS. Bronchopulmonary aspergillosis: a review and a report of eight new cases.Thorax 1952;73:317-33. 2. Meams M, Longbottom .I, Batten J. Precipitating antibodies to Aspergillus fumigatus in cystic fibrosis. Lancet 1967;1:53. 3. Brueton MJ, Grmerod LP, Shah KJ, Anderson CM. Allergic bronchopulmonaryaspergillosiscomplicating cystic fibrosis in childhood. Arch Dis Child 1980;55:348. 4. Batten JC. Allergic aspergillosis in cystic fibrosis. Mod Probl Pediatr 1967;10:227-32. 5. Laufer P, Fink VN, Bruns WT, et al. Allergic bronchopulmonary aspergillosis in cystic fibrosis. J ALLERGYCLIN IMMIJNOL1984;73:44-8. 6. Nelson LA, Callerame ML, Schwartz RI-I. Aspergillosis and atopy in cystic fibrosis. Am Rev Respir Dis 1979;120:863-78. 7. Slavin RG. Allergic bronchopulmonaryaspergillosis.In: Cher-

J. ALLERGY

CLIN. IMMUNOL. SEPTEMBER 1991

niack RM, ed. Current therapy of respiratory disease.Toronto: RC Decker, 1989:147-50. 8. Tietz NW. Gastric pancreaticand intestinal function. In: Tietz NW, ed. Fundamentalsof clinical chemistry. Philadelphia:WB Saunders, 1982:1094-9. 9. Rosenberg M, Patterson R, Mintzer R, et al. Clinical and immunologic criteria for the diagnosisof allergic bronchopulmonary aspergillosis. Ann Intern Med 1977;86:405. 10. GreenbergerPA, PattersonR. Application of enzyme-linked immunosorbentassay (ELISA) in diagnosis of allergic bronchopulmonary aspergillosis. J Lab Clin Med 1982;99:288-93. 11. Bhalla RB, Rappaport I, DeFilippi I, et al. Serum IgE levels in a New England United States Caucasianpopulation. In: BensonAA, ed. Advanced interpretationof clinical laboratory data. New York: Marcel Dekker, 1982:295-305. 12. GreenbergerPA. Allergic bronchopulmonaryaspergillosis. In: Middlteon E Jr, ReedCE, Ellis EF, Adkinson NF Jr, Yunginger JW, eds. Allergy: principles and practice. 3rd ed. St. Louis: CV Mosby, chap 51, 1988;1219-36. 13. SafirsteinBH, D’Souza MF, Simon G, et al. Five-year followup of allergic bronchopulmonaryaspergillosis. Am Rev Respir Dis 1973;108:450-9.