biphasic and protracted anaphylaxis

Biphasic and protracted anaplhylaxis

Barbara J. Stark, M.D., and Timothy J. Sullivan,, M.D. Dallas, Texas

We performed a prospective study of anaphylaxis in 25 consecutive patients. Three distinct clinical patterns were observed: uniphasic, biphasic, and protracted anaphylaxis. Thirteen patients 1’52%) experienced a single episode. Biphasic anaph$zxis occurred in jive patients (20%), two episodes of hypotension or laryngeal edema separated by asymptomatic intervals qf I to 8 hours. Initial therapy included large doses of g&o,-orticoids in three of the jive patients. Seven patients (28%) suffered hypotension, lower respiratory obstruction, or laqngeal obstructron that persisted 5 to 32 hours despite vigorous tnerap? that included systemic glucocorticoids. Recurrent or prolonged reactions were 2 .S-Jbld more likely if the onset was 30 or more minutes after exposure to the stimulus or if the offending agent had been administered by moubh (p < 0.03). Fatal reactions occurred in one patierrt who was taking propranolol und in one who was adrenal insufjicient. The results of this .itudy demonstrate that biphasic and protracted anaphylaxis are common, despite glucorticoid therapy. Potentially life-endangering recurrences occurred in jive (28%) of the 18 patients who responded to initial therapy. These observations indicate that patients should be followed care@ly after apparent remission of anaph.ylaxis. (J ALLERGY CLIN~MMUNOL 78:76-83, 1986.)

In clinical medicine the term “anaphylaxis” refers to fatal or life-threatening syndromes resulting from the sudden release of large amounts of mediators from mast cells.’ ’ The clinical picture most often reported is one of rapidly evolving respiratory obstruction or cardiovascular collapse. Approximately one in every 2700 hospitalized patients experiences drug-induced anaphylaxis.3 An estimated 400 to 800 patients die each year from allergic reactions to p-lactam antibiotics alone.4 Anaphylaxis induced by insect stings, foods, vaccines, and other drugs brings the lifetime risk to nearly 1 L;/o, 4

Although the initial clinical characteristics of human anaphylaxis are well documented,” 2.4 relatively little is known about the subsequent course of the response. Experimental studies of IgE-mediated reactions in human lung, nose, and skin indicate that strong antigen stimulation often leads to biphasic responses, with the second phase peaking 4 to 8 hours after the stimulus.‘-* In keeping with these observations, Popa and Lemer’ recently reported three cases of biphasic anaphylaxis in which bronchospasm re-

From the Departments of Internal Medicine and Microbiology, Uni- versity of Texas Health Science Center at Dallas. Dallas, Texas.

Received for publication Sept. 9, 1985. Accepted for publication Jan. 9, 1986. Reprint requests: Tirnothy J. Sullivan, M.D., Int. Med./Allergy,

University of Texas Health Science Center at Dallas, 5323 Harry Hines Blvd., Dallas. Texas 75235.

curred 3 to 4 hours after the initial manifestations. Possible recurrences 12 to 24 hours after the initial event have been mentioned but not documented.‘, I0 Systemic glucocorticoid therapy has been reported to be capable of suppressing experimentally induced second phase reactions in the lung, nose, and skin but has no known influence on initial phase reactions.. “-‘I The incidence of biphasic anaphylactic reactions and the impact of glucocorticoids on the second phase of anaphylaxis have not been resolved.

In addition to uniphasic and biphasic anaphylaxis, protracted anaphylaxis has been observed. Protracted anaphylaxis in experimental animals usually consists of a slowly developing, irreversible hypotension.14 In a review of 28 reported cases of anaphylaxis caused by oral penicillin, Batsonls noted four instances of prolonged hypotension lasting up to 18 hours. Kern and Wimberly16 described a patient whose systolic blood pressure rose to 92 mm Hg only after 11 hours of therapy. Lackey and Bukantzi7 described a patient who remained hypotensive for 2 days after an insect sting. The incidence of these prolonged responses is unknown.

In the present study, we performed a prospective analysis of the causes, presenting characteristics, and subsequent courses of 25 patients with anaphylaxis. Particular emphasis was placed on determining the incidence of recurrent or prolonged anaphylaxis and on identifying factors that might predict or diminish these complications.

76

VOLUME 78 NUMBER 1. PAW 1

Biphasic and protracted anaphylaxis 77

METHODS Patient selection

Patients observed between July 1982 and June 1984 were included in the study if they had experienced anaphylaxis and if complete clinical and therapeutic data were available. The diagnosis of anaphylaxis was based on two criteria: (1) the presence of an acute, otherwise unexplained syndrome that included hypotension, laryngeal edema, or lower respiratory obstruction and (2) clinical or immunologic phenomena that supported the diagnosis: the concurrent presence of other symptoms or signs of mast cell-mediator release such as flushing, urticaria, angioedema, or intense pruritus or evidence of the presence of IgE to the substance considered likely to have caused the reaction. Patients were excluded from this study if their illnesses were not likely to have been caused by mast cell activation or if their courses or treatment could not be verified. Each patient was examined and followed by one of the investigators at some point in the illness.

Therapeutic interventions

Initial treatment ‘was delivered by the Parkland Memorial Hospital Internal h4edicine house staff or by one of the investigators. Cardiac monitoring, airway management, oxygen, epmephrine, diphenhydramine, cimetidine, theophylline, infused sympathomimetics, and normal saline were administered, in most instances, according to published guidelines.‘8 Multiple doses of epinephrine and diphenhydramine were administered at conventional intervals until vital signs normalized and symptoms resolved. Most patients received large doses of oral or intravenous corticosteroids within 1 hour of their arrival in the emergency room. A second do,se was administered after 6 hours, and diphenhydramine was administered every 6 hours for the following 24 to 72 hours. Patients were observed carefully for 12 hours, until l:he reaction ceased if the disorder persisted longer than 12 hours. or until they expired.

Immunologic investigation

When the probable cause of anaphylaxis was an IgE- mediated reaction. attempts were made to demonstrate specific IgE by immediate wheal-and-flare skin testing. Stan- dard prick and intradermal methods, described in detail else- where,“’ were used. Patients were tested for sensitivity to penicillin, cephalosporin, insulin, equine antiserum, and selected foods

RESULTS Patient selection

A total of 35 cases were considered for entry into this inquiry. Courses and treatment could not be ad- equately verified in six patients. Two with recurrent idiopathic anaphylaxis had medicated themselves at home. One patient with hypotension was believed not to have had anaphylaxis, and another with bronchospasm and urticaria was excluded because she had

chronic asthma and had not unequivocally experienced anaphylaxis. These 10 patients were not included. yielding 25 subjects who met the criteria for entry into this study.

Causes of anaphylaxis

Twenty (80%) of the instances of anaphylaxis ap- peared to have been induced by therapeutic or diagnostic drugs (Table I). Twelve of the 25 patients (48%) reacted to antimicrobial agents, and an additional five patients, (20%) reacted to other therapeutic drugs. Three patients (12%) developed anaphylaxis during intravenous pyelography. Oral medications were im- plicated in nine patients (36%); eight had taken the drugs at home. In 11 instances (44%), anaphylaxis resulted from parenteral administration of a diagnostic or therapeutic agent; nine of these patients were in the hospital building at the onset of symptoms.

Evidence of an IgE mechanism was detected in 13 patients. Skin tests were positive in 10 of 11 patients with andphylaxis apparently caused by penicillins or cephalosporin but could not be performed in the other patient [No. 20) because of persistent antihistamine and a-adrenergic agonist therapy. The patients who reacted to insulin and antivenom also were skin test positive. One apparently food-allergic subject had a positive skin test response to soy bean extract and had been eating a soy-containing food when she became ill. No evidence of an IgE mechanism was established in the two other apparently food-allergic patients. At the preslent time diagnostic tests for reactivity to radiocontrast media, nonsteroidal anti-inflammatory drugs, and sulfonamides are not available.

Patient characteristics

The 18 women and seven men investigated are described in Tables II to IV. Patients ranged in age from 17 to 71 years with an average of 4 1 and a median of 38 years. Factors likely to have increased the risk of severe anaphylaxis’“. *’ were identified in four subjects: three were taking P-adrenergic blocking agents and one was adrenal insufficient. Patients 6: 23, and 24 had taken propranolol the day that they experienced anaphyla.xis. Patient 6 was a renal-allograft recipient who was being treated for hypertension; patient 24 was being treated for hypertension; patient 23 was receiving prophylaxis for migraine headaches. Patient 6 responded sluggishly to therapy. Multiple doses of epinephrine during 2 hours were required to treat her hoarseness and bronchospasm, despite the fact that she had no history of asthma. She had no recurrence of symptoms once her reaction subsided. After sta- bilizing slowly, patient 23 remained intermittently hy-

78 Stark and Sullivan J. ALLERGY CLIN. IMMUNOL.

JULY 1986

TABLE I. Causes of anaphylaxis in 25 consecutive patients

Agent Patients

Antimicrobial drugs (n = 12) Penicillins 7 Cephalosporins 4 Sulfamethoxazole/trimethoprim 1

Radiocontrast media 3 Foods 3 Heterologous antibodies (n = 2)

Equine antivenom 1 Murine monoclonal antihuman T3 1

Aspirin 1 Zomepirac 1 Insulin 1 Unknown 2

potensive for 12 hours. Patient 24 developed intract- able upper and lower respiratory obstruction after injection of monoclonal antibody for suppression of allograft rejection and could not be ventilated effec- tively for several minutes, even after intubation. The prolonged hypoxia caused a severe cerebral injury that led to his demise weeks later.

The adrenal-insufficient patient (No. 20) also expired. At autopsy virtually no adrenal cortical or med- ullary tissue could be found in her calcified adrenal remnants. Since the patient had been treated for pulmonary tuberculosis several years before her death, tuberculosis was presumed to have caused the destruc- tion of her adrenal glands. Her failure to seek medical attention during the first 18 hours of anaphylaxis was believed to have contributed to her poor response to therapy after admission to the hospital.

Characteristics of the anaphylaxis

The life-endangering manifestations of anaphylaxis detected are summarized in Table VI. Other symptoms and signs of anaphylaxis such as pruritus, flushing, angioedema, urticaria, gastrointestinal disturbances, and metallic taste in the mouth are not recorded in the tables, but they were often present.

In 11 patients (44%) symptoms began within 30 minutes of the time the patients were exposed to the apparent cause of the anaphylaxis. In 10 patients (40%) the onset was delayed 30 to 60 minutes, and in two other patients the reactions began after a 2- to 3-hour asymptomatic interval. In the remaining two patients the interval could not be determined because the causes were unclear.

In biphasic reactions the second episode always included the same major signs and symptoms as the first. One patient (No. 13) had a recurrence of her

urticaria, but since she did not experience hypotension or respiratory obstruction in the second phase, she was not included in the biphasic anaphylaxis group.

Similarities and differences among patient groups

In general the symptoms and signs noted and therapy administered to the patients were similar, regardless of the clinical course that ensued (Tables 11 to V). Laryngeal edema occurred more frequently in the protracted (57%) and biphasic (40%) groups than in the uniphasic group (23%), but these differences were not statistically significant. Epinephrine and diphen- hydraminle were administered to 92% and 88% of the patients, respectively. Corticosteroids were administered to 19 (76%) of the patients, and cimetidine was administered to 15 (60%). Six of the 13 patients with bronchospasm (46%) required theophylline in addition to inhaled and injected P-adrenergic agonists.

Anaphylaxis provoked by an oral agent was significantly more likely to be complicated by recurrent or prolonged reactions: nine of 12 versus three of 11; p < 0.03 by chi-square and Fisher exact tests; relative risk 2.8. Of the 11 patients who had received parenteral stj muli, eight (73%) had uniphasic, two ( 18%) had biphasic, and 1 patient (9%) had protracted anaphylaxis. Orally administered agents induced uniphasic anaphylaxis in three of 12 (25%), induced biphasic in three (25%), and induced protracted in six (50%).

Anaphylaxis that began 30 or more minutes after exposure to the stimulus also was significantly more likely to be complicated by recurrent or prolonged reactions: nine of 12 versus three of 11; p < 0.03 by chi-square and Fisher exact tests; relative risk 2.8. Of the 11 patients whose reactions began within 30 minutes, eight had uniphasic, two had biphasic, and one patient h,ad protracted anaphylaxis. Of the 10 patients whose reactions began 30 to 60 minutes after the stimulus, three had uniphasic, two had biphasic, and five patients had protracted anaphylaxis. Two patients had onsets more than an hour after exposure to the stimulus; one experienced a biphasic response, and one experienced a protracted response.

DISCUSSION

Recent studies have provided major advances in our understanding of the immunologic, cellular, and bio- chemical basis of acute allergic reactions. ‘. ’ In marked contrast, direct studies of clinical anaphylaxis have been few and have involved small numbers of patients.‘. ‘2. 23 The paucity of systematically acquired, first-hand information can be attributed in part to the rarity, unpredictability, and rapid evolution of the re-

VOLUME 78 NUMBER 1. PART 1


TABLE II. Characteristics of patients with uniphasic anaphylexis --

Therapyt Pt -

No. Age (yr)/sex Agent/route Signs* E H, H, Steroid Theo

I 47/M

8 42/M 9 66,‘F

10 34,‘F 11 48,‘F

12 71iF 13 25/F

57/F 17/F 26lF 34/F 27,‘M 55/F

Penicillin IV Penicillin IM Penicillin SC Penicillin PO Cephazolin IV RadiocontrastS

media Radiocontrast

media Antivenom Insulin Food 1 hod

Food Unknown

H: O/O, CA H: SO/50 B L, B H: 60/O L, B

B, RA, H: O/O

B, H: 70/O B, H: 80/O B, H: 60/O H: 90160

(before 160/80) 1

L

H: 90170 (after E)

SC IV SC IM SC IV IV IV IV

IV SC IV IV

IV IV IV

INF IV IV SC IV SC IM SC IV IV

SC IV IV SC IV IV

2OOmgC 60 mg P 50 mg P

IV

200 mg C IV

200mgC

12.5 mg C 60 mg P

125 mg C 60 mg P

I’t = patient; IV = mtravenously; SC = subcutaneously; IM = intramuscularly; FQ = by mouth; INF = infusion. *H, hypotension; CA, cardiac arrest; B, bronchial obstruction; L, laryngeal obstruction; RA, respiratory arrest. tE, epinephrine (SC 0.3 to 0.5 mg, IV 0.3 to 0.5 mg, INF 2 pg/min IV); H,, diphenhydramine (50 to 100 mg); H,, cimetidine (300 mg);

steroid: C, hydrocortisone; P, methylprednisone; Theophylline, aminophylline (250 to 500 mg). *Patient taking propranolol.

TABLE III. Characteristics of patients with biphasic anaphylaxis

Therapyt Pt -

No. Age (yrbex Agentlroute Signs* E H, Hz Steroid Theo

14 28/F

15 24/M

16 67/M

17 34iF

18 21iF

Penicillin IV

Penicillin PO

Cephalexin PO

Cephalexin PO

Radiocontrast media

1 B, H: O/O 2 H: 80/50 (8 hr) 1 L, B, H: 100/O 2 L, B, RA (1 hr)

I H: 8010 2 H: 80150 (4 hr) 1 L 2 L (6 hr) 1 H: 40/O 2 H: 80150 (2.5 hr)

IV IV IV 125 mg P (fluids only)

IV (V tach) IM 500 mg C

(intubated) IV

(fluids only) SC IV IV 125 mg C SC IV IV 60 mg P SC IM 12.5 mg C

(fluids only)

IV

IV

*See Table II. The numbers 1 and 2 refer to the initial and second phase reactions. The asymptomatic interval between phases is presented in parentheses

tSee Table II.

actions. This study demonstrates that university hos- into anaphylaxis inaccessible to previous studies that pitals and their emergency facilities can provide a have relied on small numbers of cases or review of significant number of patients with anaphylaxis in a published accounts of selected cases. coordinated context appropriate for clinical investi- Apprortimately half of the patients presented in this gation. In additiorn, this study of 25 consecutive pa- article (12 of 25) experienced courses that were com- Gents treated in a relatively consistent manner, and plicated by recurrent or refractory anaphylaxis. The observed by the investigators, has provided insight patients received similar treatment for their initial ep-

80 Stark and Sullivan J. ALLERGY Cl-IN. IMMUNOL. JULY 1986

TABLE IV. Characteristics of patients with protracted anaphylaxis ---

Therapyt Pt

No. Age (yrllsex Agent/route Signs* E H, H, Steroid Theo Duration

19 57/F Penicillin PO H, B, L. CA IV IV 200 mg C IV B: 32 hr 20 33iF Cefaclor PO H. CA LNF IV IV 1000 mg C H: 8 days*

Expired 21 42 F Suifa/Tmp PO H: 80140 SC IV IV 125 mg P H: 24 hr 22 50’F Aspirin PO L. SC IV IV 60 mg P L: 18 hr 23 34:F; Zomepirac POP B, H: 7010 SC 1V IV 100mg C IV H: 12 hr 24 501 M Murine monoclonal L, B, RA SC 500 mg C B: 5 hr

anti-T3 anti- Expired body3

25 38:M Unknown L, B SC IV IV 500 mg C L, B: 6 hr

Sulfaitmp = sulfamethoxazoleitrimethoprim. *See Table II. t.See Table II. $This patient received infusions of epinephrine and norepinephrine during 8 days. fdl other medications were administered repeatedly at

conventional invervals throughout her course. PThese patients were taking propranolol.

isodes (Table V) roughly in keeping with recom- mended protocols. I’ *. lo. ‘* Thus inadequate or incon- sistent therapy of the presenting syndromes does not appear to account for the persistence or recurrence of anaphylaxis. In addition, patients’ subsequent courses could not be predicted from the presence or severity of hypotension or respiratory obstruction at the time of initial evaluation (Table V). Although irretrievably moribund cases and instances of mild anaphylaxis were few in this series, the data support the conclusion that recurrent or prolonged anaphylaxis is common.

The 48% incidence of late sequelae detected in this study is higher than has been appreciated from in- spection of published accounts of selected case his- tories. l-4, *‘-x The small number of cases in most re- ports and the absence of direct observation of patients after apparent remission may account for the failure of earlier studies to recognize a high incidence of these complications.

Recurrent and prolonged anaphylaxis were 2.8-fold more common in patients who had delayed onsets of reactions and in patients who reacted to orally administered agents. In some reported instances, ulti- mately fatal anaphylaxis had begun several hours after exposure to an antigen.25,27 This has led to speculation that patients with delayed onsets may be at greater risk of prolonged anaphylaxis and that they may be at greater risk of a fatal outcome. Our data support the concept that delayed onset anaphylaxis is more likely to be complicated by recurrent or refractory manifestations and suggest that the degree of this increased risk is approximately 2.8-fold. Our data also

suggest that the customary 30-minute period of observation iafter drug administration, although it is use- ful, would not be long enough to detect the onset of many of the more complicated and refractory cases. These apparent risk factors must be examined in new hypothesis-testing studies to assess rigorously their predictive value and their implications.

Anaphylaxis has been fatal in approximately 3% to 9% of re.ported cases.4. I4 Of these fatal reactions, death occurred within an hour of the onset of symptoms in 55% to 80% of the cases.25, 26 Two patients in the present series (8%) died despite vigorous therapy, but both deaths occurred more than a week after the initial manifestations of anaphylaxis. One patient was taking propranolol, an important risk factor for anaphylaxis.*’ The other, a previously undiagnosed adrenal-insufficient patient, came to medical attention 18 hours after her reaction began. One of our patients with biphasic anaphylaxis (No. 15) almost certainly would have died if he had not been under observation when his airway obstruction recurred.

Three of our patients experienced recurrent hypotension, one had recurrent laryngeal and bronchial obstruction (No. IS), and one had recurrent laryngeal edema. Late-phase reactions in skin and lung occur in patients with high specific IgE levels or after un- usually large doses of antigen,‘. 7. I’, ** raising the pos- sibility that our patients had biphasic immunologic reactions. Alternatively, the phenomena could reflect waning pharmacologic suppression of an ongoing mediator-release process. The prolonged asymptomatic intervals noted in several patients favor the former

VOLUME 78 NUMBER 1, PART 1


TABLE V. Presenting signs and initial therapy: Uniphasic, biphasic, and protracted anaphylaxis

No. of patients (% of group)

Total 25

Uniphasic: 13

Biphasic 5

Protracted 7

Signs Hypotension Bronchial obstruction Laryngeal ‘edema

Therapy Epinephrin’e Diphenhydramine Cimetidine Glucocorticoids Theophylline

17 (68) 13 (52) 9 (36)

23 (92) 22 (88) 15 (60) 19 (76)

6 (24)

9 (69) 7 (54) 3 (23)

12 (92) 12 (92) 8 (62) 9 (69) 2 (15)

4 (80) 2 (40) 2 (40)

4 (80) 4 (80) 2 (40) 3 (60) 1 cm

explanation, but we did not collect data that would permit definition of the roles of these factors. Subject NO. 15 required endotracheal intubation for a respiratory arrest during the second phase. This patient had an asymptomatic interval of approximately 1 hour before his recurrence, highlighting the difficulty of distinguishing immunologic from pharmacologic ex- planations. Regardless of the mechanisms involved. however, our data indicate that potentially life-threatening recurrent anaphylactic reactions occur in one of five patients despite conventional therapy.

Current recommendations for the management of anaphylaxis are based on knowledge of the patho- physiologic mechanisms involved, results of experi- ments in animal models, and published anecdotes. No controlled clinical studies of the effectiveness of treatment for this syndrome have been published. Endo- tracheal intubation or cricothyrotomy for severe upper airway obstruction need no validation of that kind, but some common pharmacologic interventions are less obviously effective. Several investigators have recom- mended that patients experiencing anaphylaxis should be treated with systemic glucocorticoids to suppress ongoing or recurrent reactions. ‘. *, lo. “. I8 Studies of individual organs have demonstrated impressive inhibition of late-phase reactions by administration of glucocorticoids before antigen challenges. Multiple 30 to 40 mg doses of prednisolone administered 12 hours to 7 days before antigen injection prevent or diminish human late-phase skin reactions.‘*. z9, j” A single dose of prednisolone immediately prechallenge has been reported to cause a 50% reduction of the 6-hour reaction diameter.” Late-phase falls in FEV, after antigen-inhalation challenge also can be suppressed by premed- ication with prednisolone.29 The biphasic pulmonary responses are associated with biphasic mast cell-mediator release ,“9 but whether glucocorticoids suppress

4 (57) 4 (57) 4 (57)

7 (100) 6 (86) 5 (71) 7 (100) 3 (43)

-

TABLE VI. Life-endangering manifestations of anaphylaxis in 25 patients

Reaction No. of patients

Hypotension Bronchial obstruction Laryngeal obstruction Cardiac arrest Respiratory arrest

17 (68%) 13 (52%) 9 (36%) 3 (12%) 3 (12%‘)

second-phase mediator release, inhibit tissue responses IO the mediators, or both is not known.

The results of this study indicate that contrary to the expectations noted above, glucocorticoid therapy introduced during the initial phase of anaphylaxis does not prevent the appearance of recurrent or protracted anaphyla.uis. Ten of the 12 patients with recurrent or protracted life-endangering reactions received high doses of either methylprednisolone or hydrocortisone, beginning shortly after the initial phase was stabilized. Two of the three patients with recurrent bronchial obstruction reported by Popa and Lerner’ had received glucocorticoids during their initial therapy. Although the drugs may not have had time to act in some cases or the suppressive force may have been overwhelmed, glucocortrcoids clearly cannot be relied on to suppress recurrent or prolonged anaphylaxis. In most instances the biphasic reactions did respond to other conventional interventions, indicating that they were not sim- ply refractory responses. These observations are consistent wil:h the notion that glucocorticoids do not inhibit second-phase mast cell-mediator release. The effectiveness of steroids in the isolated skin and lung studies may reflect impact on cellular inflammation that is not as important or as susceptible to glucocorticoid inhibition in anaphylaxis.

82 Stark and Sullivan J. ALLERGY CLIN. IMMUNOL. JULY 1986

Increasing evidence indicates that antihistamines may have a place in the therapy of initial and second phases of anaphylaxis. Combined H, and H2 receptor blockade has been demonstrated to abolish late-phase cutaneous responses,3’. 32 and H, blockers alone have been demonstrated to decrease immediate cutaneous reactions in both experimental and clinical settings.3’. 3’ SchGning et a1.22 reported that premedi- cation with combined H, and H2 blockers suppresses clinical manifestations of chemically induced anaphylaxis in both dogs and humans. Pretreatment with antihistamines and corticosteroids has become standard for suppression of reactions to radiocontrast media.34 Combined H, and H2 blockade might contribute to the suppression of recurrent or prolonged anaphylaxis, but this issue must be examined in controlled studies.

The results of this investigation lead to two partic- ularly important conclusions about clinical anaphylaxis. First, life-endangering manifestations recur in approximately 20% of cases and can appear up to 8 hours after apparent remission of anaphylaxis. Sec- ond, glucocorticoids do not reliably prevent recurrent or prolonged anaphylaxis and may not exert any effect at all. Current approaches to anaphylaxis that include discharge of patients soon after apparent remission and those that rely on glucocorticoid suppression of prolonged or recurrent anaphylaxis may be danger- ously inappropriate. Until methods are developed to predict or avoid late-phase anaphylactic reactions, all patients should be observed for several hours after apparent recovery from acute anaphylaxis.

REFERENCES

1. Wasserman SI: Anaphylaxis. In Middleton E, Reed CE, Ellis EF, editors: Allergy: principles and practice. St. Louis, 1983, The CV Mosby Co, pp 689-99

2. Sheffer AL: Anaphylaxis. J ALLERGY CLIN IMMUNOL 75:227, 1985

3. Porter J, Jick H: Drug-induced anaphylaxis, convulsions, deaf- ness, and extrapyramidal symptoms. Lancet 1:587, 1977

4. Valentine M, Frank M, Friedlander S, et al: Allergic emergencies. In Krause RM, editor: Asthma and the other allergic diseases. IY79, NIAID Task Force Report, National Institutes of Health, pp 467-507

5. Robertson DG, Kerigan AT, Hargreave FE, Chalmers R, Do- lovich J: Late asthmatic responses induced by ragweed-pollen allergen. J ALLERG’< CLIN IMMUNOL 54:244, 1974

6. Slavin RG, Fink JN, Becker RJ, Tennebaum JI, Feinberg SM: Delayed response to antigen challenge in induced delayed reactivity. J ALLERGY 35:499, 1964

7. Solley GO. Gleich GJ, Jordon RE, Achroeter AL: The late phase of the immediate wheal and flare skin reaction. J Clin Invest 58:408, 1976

8. Price JF, Hey EN, Soothill JF: Antigen provocation to the skin, nose, and lung in children with asthma: immediate and dual hypersensitivity reactions. Clin Exp Immunol 47587, 1982

9. Popa VT, Lemer SA: Biphasic systemic anaphylactic reaction: three illustrative cases. Ann Allergy 53:151, 1984

10. Bleecker ER, Lichtenstein LM: Systemic anaphylaxis. In Lich- tenstein LM, Fauci AS, editors: Current therapy in allergy and immunology 1983-84. Philadelphia, 1983, BC Decker Inc, pp 78-83

11. Zetters:riim 0: Dual skin test reactions and serum antibodies to subtilisin and Aspergillus fumi,qatus extracts. Clin Allergy 8:77, 1978

12. Poothullil J, Umemoto L, Dolovich J, Hargreave FE, Day RP: Inhibition by prednisone of late cutaneous allergic responses induced by antiserum to human IgE. J ALLERGY CLIN IMMUNOI. 57:164, 1976

13. Griinn’:berg R, Strandberg K, Stalenheim G, Zetterstriim 0: Effect in man of anti-allergic drugs on the immediate and late- phase cutaneous allergic reactions induced by anti-IgE. Allergy 36:201, 198 I

14. Becke- EL, Austen KF: Anaphylaxis. In Miescher PA, Muller- Eberhard HJ, editors: Textbook of immunopathology. New York, 1968 Grune & Stratton Inc, pp 76-93

15. Batson JM: Anaphylactoid reactions to oral administration of penicillin. N Engl J Med 262:590, 1960

16. Kern RA, Wimberly NA: Penicillin reactions: their nature, growing importance, recognition, management, and preven- tion. Am J Med Sci 226:357, 1953

17. Lackey RF, Bukantz SC: Allergic emergencies. Med Clin North Am 58:147, 1974

18. Sullivan TJ, Stark BJ: Drug reactions. In Lichtenstein LM, Fauci AS, editors: Current therapy in allergy and immunology 1983-84. Philadelphia, 1983, BC Decker Inc. pp 62-9

19. Wendcl GD, Stark BJ, Jamison RB, Molina RD, Sullivan TJ: Penicillin allergy and desensitization in serious infections during pa:gnancy. N Engl J Med 312:1229, 1985

20. Ansell G, Tweedic MCK, West CR, Evans P, Couch L: The current status of reactions to intravenous contrast media. lnvest Radio 15(suppl):32, 1980

21. Jacob:. RL, Rake GW, Foumier DC, Chilton RJ, Culver WG, Beckman CH: Potentiated anaphylaxis in patients with drug- induced beta-adrenergic blockade. J ALLERGY C[.~rj IMMUNOL 68:125, 1981

22. Schanlng B, Lorenz W, Doenicke A: Prophylaxis of anaphy- lactoitl reactions to a polypeptidal plasma substitute by H,- plus H2-receptor antagonists: synopsis of three randomized controlled trials. Klin Wochenschr 60:1048, 1982

23. Smith PL, Kagey-Sobotka A, Bleecker ER, et al: Physiologic manifestations of human anaphylaxis. J Clin Invest 66: 1072, 1980

24. Welch H, Lewis CN, Weinstein HI, Boeckman BB: Severe reactions to antibiotics: a nationwide survey. In Welch H, Marti-Ibanez, editors: Antibiotics annual 1957-58. Proced- dings of the Fifth Annual Symposium on Antibiotics. New York, 1958, Medical Encyclopedia, pp 296-309

25. Idse 0, Guthe T, Willcox RR, deWeck AL: Nature and extent of pericillin side reactions, with particular reference to fatal- ities from anaphylactic shock. Bull WHO 38:159, 1968

26. Delag: C, Irey NS: Anaphylactic deaths: a clinicopathologic study of 43 cases. J Forensic Sci 17:525, 1972

27. Bernard JH: Studies of 400 Hymenoptera sting deaths in the United States. J ALLERGY CLIN IMMUN~L 52:259. 1973

28. Gleich GJ: The late-phase of immunoglobulin E-mediated reaction: a link between anaphylaxis and common allergic dis- ease’? J ALLERGY CLIN IMMUNOL 70:160, 1982

29. Booij.Noord H. DeVries K, Sluiter HJ, Orie NGM: Immediate and late bronchial obstructive reactions to inhalation of house dust and protective effects of disodium cromoglycate and prednisolone. J ALLERGY CLIN IMMUNOL 48:344. 1971

30. Dor PI, Vervloet D, Sapene M, Andrdc L. Bonerandi JJ, Char-

VOLUME 78 NUMBER 1, PART ‘I

pin J: Induction of late cutaneous reaction by kallikrein injection: comparison with allergic-like late response to compound 48180. J ALLERGY CLIN IMMUNOL 71:363, 1983

31. Smith JA. Mansfield LE, deShazo RD, Nelson HS: An evaluation of the pharmacologic inhibition of the immediate and late cutaneous reaction to allergen. I ALLERGY CLIN IMMUNOL 65:118, 1980

32. Lemanske R, Barr L, Guthman D, Kaliner M: The effects of antihistamines on rat cutaneous early and late phase reactions. J ALLERGY CLIN IMMUNOL 71:136, 1983 (abst)

Biphasic and protracted anaphylaxis

33. Mathews KP, Hemphill FM, Love11 RG, Forsythe WE, Sheldon JM: A controlled study on the use of parenteral and oral antihistamines in preventing penicillin reactions. J ALLERGY 27: I, 1956

34. Greenxrger P, Patterson R, Kelly J, Stevenson DD, Simon R, Lieberman P: Administration of radiographic contrast media in high-risk patients. Invest Radio1 15(suppl~:40, 1980

Decreased skin response to intradermal histamine in cancer patients

Claude Burtin, Ph.D.,* Christian Noirot, M.D.,** Christian Giroux, M.D.,*** and Pierre Scheinmann, M.D.* Paris, Ferolles Attilly, and Bobigny, France

The skin response to intradermul injections of serial dilutions of histamine was studied in 68 cancer patients and 29 noncancer, nonallergic subjects. In cancer patients with primary tumor with or without metastasis, incidence and size of wheals and flare were markedly depressed by com>oarison with control subjects. Pseudopods were rare, and itching was absent. Similar results were observed in patients in which excisions of the primary tumor dated from <5 weeks. In contrast, when excision dated from >2 months, skin test:: returned to almost normal values. Intermediate values were obtained in patients with metastases alone. This decrease in skin response to histamine was not due to a peripheral neuropathy in cancer patients, since results of electrophysiologic studies were not dtfterent in cancer and noncancer patients. The presence of

a tumo,’ mimics the effects of general administration of histamine H, antagonists on the skin response to histamine. (J ALLERGY CLIN IMMUNOL 78:83-9, 1986.)

Clinical data have demonstrated an opposition between allergy and cancer. In patients with clinical manifestations of a.topy, the incidence of cancer is less frequent than in healthy subjects. By contrast, the incidence of atopy is more frequent in noncancer patients than in cancer patients, even in those in which immunologic responses are normal.‘” These results

From the *Facultt de Mtdecine Necker-Enfants Malades, Paris, **Centre Medical de Recherches et de Traitements Diet& tiques de Forcilles, Ferolles Attilly, and ***Service d’Explora- tions Fonctionnelles (Physiologie), HBpital Avicenne, Bobigny, France.

Supported by grants from the Association pour la Recherche sur le Cancer and Ligue dipartementale contre le Cancer de Seine et Marne.

Received for publication Aug. 2, 1985. Accepted for publication Jan. 9, 1986. Reprint requests: Claude Burtin, Ph.D., U 200, FacultC de

Mtdecine Necker-Enfants Malades, 156 rue de Vaugirard, 75730 Paris Cedex 15, France.

have nol: been observed by some authors, but their criteria of atopy were questionable.‘-’

Indeed, the relationship between allergy and cancer remains controversial and was supported by Lynch et al.‘” but not by Rosenbaum and Dwyer.”

We have previously demonstrated that blood histamine levels are significantly lower in patients with primary tumor with or without metastasis and in patients with metastasis alone than in noncancer patients. Moreover, a progressive decrease in blood histamine levels preceded clinical relapse and/or detection of metastasis. In contrast, after successful tumor rejection, blood histamine levels returned to nearly normal levels. ‘* Since patients with advanced cancer tolerate large doses of parenteral histamine,13 we attempted to study the sensitivity of their histamine receptors by analysis of the skin response to intradermal histamine injections and comparison with healthy subjects and noncancer patients. As the flare component of Lewis’

83

biphasic and protracted anaphylaxis

Documents