Portal de Periódicos da CAPES

Allergy skin testing in predicting adverse reactions to fluorescein: a prospective clinical study

2009; Wiley; Volume: 89; Issue: 5 Linguagem: Inglês

10.1111/j.1755-3768.2009.01722.x

1755-3768

D. Kalogeromitros, Μichael Μakris, Xenophon Aggelides, Anagnostis I. Mellios, Fani Giannoula, Kyriaki Sideri, Alexander Rouvas, Panagiotis Theodossiadis,

Food Allergy and Anaphylaxis Research

Purpose: To evaluate allergy skin testing as a diagnostic tool of adverse reactions to fluorescein and whether allergy and previous sodium fluorescein angiography (SFA) act as predisposing factors. Methods: Patients with adequate indication for fluorescein angiography and normal skin responsiveness were subjected to allergy skin-prick and intradermal tests for fluorescein, followed by SFA. During SFA, adverse reactions were monitored and classified as mild, moderate or severe. Previous SFAs and adverse reactions as well as the presence of atopy were also registered. Results: One thousand and thirty-seven patients were enrolled in the study and 1284 SFAs were executed. Forty-four patients (4.3%) developed 55 adverse reactions; among them 50 (3.8%) were mild, three (0.2%) moderate and two (0.16%) severe. None of the reactors produced positive skin tests to fluorescein. Patients with atopy and previous SFAs were not more susceptible to adverse reactions. Conclusion: The vast majority of adverse reactions to fluorescein are mild and not attributed to immunological mechanisms. Allergy skin tests cannot predict non-immunological reactions but their utility remains substantial in predicting anaphylaxis during SFAs and must be performed in patients reporting risk factors in their past medical history. Sodium fluorescein angiography (SFA) is considered to be the optimum diagnostic method for the assessment of chorioretinal diseases, such as retinal complications of diabetes mellitus, retinal vein occlusion, age-related macular degeneration, etc. The increasing number of performed SFAs is well demonstrated by data acquired from the USA. In 2006, more than 1 300 000 SFAs (Centers for Medicare and Medicaid Services 2006) were executed while their number in 1984 was estimated at over 650 000 (Yannuzzi et al. 1986). These proportions reflect the high levels of safety and reliability of this procedure. Safety in particular has been established by large retrospective studies, showing a low prevalence of serious adverse reactions (ARs). Among them, anaphylaxis is a scarcely reported but potentially lethal complication (Fineschi et al. 1999) and may pose significant danger considering the extended use of SFA worldwide. Anaphylaxis to intravenous fluorescein can be confirmed easily by means of allergy skin testing (Trindade-Porto et al. 1999; Nucera et al. 2003). Our study was designed primarily to estimate the ability of fluorescein skin-prick and intradermal tests to predict SFA reactions. Furthermore, we recorded reactions during the SFA procedure in a large, well-characterized patient population. This study was conducted with the collaboration of the Allergy Unit and the Department of Ophthalmology of Attikon University Hospital in Athens, Greece. Patients referred to the Ophthalmology Department during a 42-month time period (January 2005–June 2008) with adequate indication for SFA were enrolled to the study. A brief medical history was obtained initially and all patients were referred to the Allergy Unit. Before any diagnostic evaluation, the personal and family history of atopic diseases (allergic rhinoconjuctivitis ± asthma, atopic dermatitis) and drug allergies were investigated. Patients were also asked about previous SFAs and any related ARs that they recalled. Whether medication was prescribed or over-the-counter was also recorded. During the 42 months of the study, 1911 patients visited the Ophthalmology Department in order to undergo an SFA. These patients were invited to participate in our survey after they had been given a full explanation of the experimental details. Sixty-two refused to sign the written consent and were therefore excluded. Seven hundred and sixty-three patients receiving antihistamines, corticosteroids in high doses, antidepressants and other psychoactive components were also excluded because these drugs compromise the integrity of allergic skin tests. Seventeen more were excluded because they did not produce adequate reactions at the positive control during the skin tests, thus skin responsiveness could not be evaluated. Thirty-two of the patients failed to show up for SFA after completing the allergy tests. The remaining 1037 patients [480 male, 557 female; mean age 65.7 years, range 15–94, standard deviation (SD) 14.26] were included in the analysis. After being submitted to allergy skin tests, these patients underwent 1284 consecutive fluorescein angiographies. According to protocol design, all patients underwent in vivo allergological evaluation before SFA. In patients undergoing more than one SFA, skin testing was repeated. Before the fluorescein administration, baseline vital signs and peak expiratory flow rate were recorded. Phenylephrine 2.5% was applied for pupil dilation. The SFA procedure involved the intravenous injection of 5 ml of sodium fluorescein 10% into the antecubital vein with an infusion rate of 1 ml/second. The time of onset and the type of ARs were documented by two retinal angiographers. During the procedure, an allergologist and an anaesthesiologist remained on call to evaluate and treat moderate or serious ARs. Resuscitation facilities were always available during the procedure. According to a previously published classification system (Chazan et al. 1971; Yannuzzi et al. 1986; Kwan et al. 2006), depending on the duration, the need for treatment and the final outcome, all the observed reactions were recorded and classified as follows: mild transient reactions not requiring any treatment (nausea and vomiting, sneezing, pruritus); moderate reactions usually requiring intervention (urticaria, syncope, thrombophlebitis, pyrexia, local tissue necrosis, nerve palsy); and severe reactions that may compromise the patient's health and may demand more urgent treatment (anaphylaxis, brochospasm, myocardial infarction, seizures). The possible relation between several factors (atopy, previous SFA procedures and previous SFA reactions) and the reaction rate was evaluated statistically using Pearson's χ2 test. Possible sensitization to fluorescein was evaluated through skin-prick and intradermal tests. For the skin-prick tests, increasing concentrations of sodium fluorescein were used successively (1/100, 1/10, 1/1 dilutions of the full-strength preparation). Intradermal tests were performed using a 1/100 dilution of fluorescein 100 mg/ml (sodium fluorescein 10%; Institute of Pharmacological Research and Technology, Pallini, Greece). Histamine chloride (10 mg/ml; Stallergenes, Paris, France) and 50% glycerin saline solution were used as positive and negative controls, respectively. Intervals of 20 min were kept between each test. Wheal and flare reactions typical of mast-cell activation were evaluated and documented by the allergologist conducting the tests. Prick tests with a wheal diameter at least 3 mm larger than the one produced by the negative control were considered positive. During the intradermal testing, any occurring wheal and flare reaction with a wheal diameter ≥ 5 mm was also considered positive. This study was performed in accordance with the tenets of the Declaration of Helsinki. In the study population, skin-prick and intradermal tests were all negative except in a 67-year-old woman with a negative history of atopy and no previous SFAs. The patient had negative skin-prick tests while intradermal tests with the 1/100 preparation produced a 2-mm-diameter wheal. Intradermal injection of the 1/10 preparation as well as 1/1 preparation were both positive (7/13 mm and 7/16 mm wheal and flare diameter, respectively). The positive skin responses were probably caused by direct (non-immunoglobulin E) mast-cell degranulation because no previous exposure to fluorescein was described. The patient had an absolute indication for SFA, so she underwent the procedure under prophylactic treatment (methylprednisolone 16 mg, levosetirizine 5 mg and ranitidine 150 mg administered at 13, 7 and 1 hr prior to the procedure) without any reaction. Forty-four patients had ARs to intravenous fluorescein (overall rate of reactors 4.3%). The total number of reactions was 55 (overall rate of reactions 4.3%) (Table 1). According to the previously mentioned classification system, the observed reactions were: 50 mild (3.9%), three moderate (0.2%) and two severe (0.16%), all in patients with negative skin tests. The mild ARs consisted of: (i) nausea and vomiting in 39 angiographies (3% incidence), usually remitting after 30 min without treatment (antiemetics were administered in two patients because of protracted nausea and vomiting); (ii) face and upper trunk flushing in six angiographies (0.5%) appearing about 30–60 min after the infusion –(with the exception of one presenting 24 hr later), all resolving within 2 hr without treatment; (iii) mild dizziness or headache in three patients (0.2%), developing immediately after the infusion in one patient and after 1 and 3 hr, respectively, in the other two; (iv) generalized pruritus in two angiographies (0.16%) presenting 15 and 90 min after the fluorescein administration, respectively, with no further progression. Three patients presented with moderate reactions: two of them with localized urticaria in the face and the extremities, respectively, commencing 5 and 35 min after the injection and lasting < 2 hr under treatment with intravenous corticosteroids and antihistamines; and the third with angioedema of the upper lip noticed 6 hr after the SFA procedure and lasting for 24 hr. Two out of these three patients had undergone SFAs in the past, both with no ARs. Severe reactions occurred in two patients (0.16% of angiographies). The first patient developed a serious anaphylactoid reaction 5 min after the injection with hypotension (60/35 mmHg), tachycardia (145 bpm), diffuse erythema and pruritus. Blood pressure returned to normal shortly after a subcutaneous injection of 0.3 mg of aqueous adrenaline 1 : 1000. Intravenous steroids and antihistamines were also administered. The patient was monitored closely for 12 more hours and then discharged. Tryptase concentration was increased in serum 2 hr after the reaction onset (17.9 μg/l with normal value < 13 μg/l; Tryptase Phadia, Uppsala, Sweden). The second patient developed nausea, diaphoresis, back pain and angina 1 hr after the injection. Electrocardiogram (ECG) and cardiac enzymes revealed an anterior myocardial infarction and the patient was admitted to the Cardiology Department for further management. The patient had no previous known history of coronary disease but was suffering from diabetes mellitus and had been receiving oral antidiabetic agents for the past 9 years. None of these two reactors reported previous SFA. The presence of atopy as well as a previous AR suggests an increased risk of ARs. However, this does not have any suggestive significance (atopy and SFA ARs, p = 0.696; previous SFAs and SFA ARs, p = 0.635). Specifically, 188 patients (18.1%) reported symptoms of atopic disease. Only seven of them experienced mild ARs. Four hundred and twelve of 1037 patients (39.7%) had undergone SFAs in the past. The rate of ARs in that subset was 19/412 patients (4.6%); 13 of these patients reported reactions in the past (mostly nausea and pruritus). Thus, the rate of recurrence was found to be 68.4%. This study was designed to evaluate allergy skin testing as a screening tool in order to predict potentially serious hypersensitivity reactions to SFAs. One thousand, two hundred and eighty-four cases of skin testing followed by SFA were assessed, with only one occasion of positive skin reaction to fluoroscein that was not associated with AR during the SFA procedure. A single occasion of positive skin reaction to fluorescein was detected and subjected to SFA with no AR. We confirm that IgE-mediated reactions to fluorescein administration are very limited; thus skin testing as a screening procedure in the general population is not indicated. However, in vivo testing with fluorescein remains the most reliable diagnostic tool for the diagnosis of IgE-mediated allergy and its most severe manifestation (anaphylaxis). In addition, previous SFAs or the presence of atopy do not affect the reaction rate. The overall incidence of ARs is estimated to be between 3% and 20% (Stein & Parker 1971; Alvis 1985; Patz et al. 1986; Singerman 1986; Yannuzzi et al. 1986; Kwiterovich et al. 1991; Kwan et al. 2006; Lira et al. 2007). This variability could be attributed to the underestimation of mild reactions, especially from large retrospective surveys. We found that rate to be 4.3%. Nausea and vomiting are the most common mild ARs, the prevalence being below 5% in most studies (Karhunen et al. 1986; Yannuzzi et al. 1986; Kwiterovich et al. 1991; McLauchlan et al. 2001). Moderate ARs consist mainly of urticaria and angioedema. Previous studies estimated the risk for moderate reactions at between 0.2% and 1% (Chazan et al. 1971; Yannuzzi et al. 1986; Kwan et al. 2006), which is in accordance with our findings. Severe reactions are scarcely reported. The most adequate data on these reactions come from two major surveys (Zografos 1983; Yannuzzi et al. 1986), which revealed a frequency of 1 : 1900 and 1 : 18 020 serious reactions, respectively. The incidence of deaths related to SFA was 1 : 220 000 and 1 : 49 557, respectively. In our study, the frequency rate of severe reactions was 2/1284 angiographies (0.16%). Previous attempts to indicate possible causative factors related to the procedure technique [e.g. alteration of concentrations (Justise et al. 1977), rate of infusion (Chazan et al. 1971), injected volume (Kwan et al. 2006)] have not been conclusive. Greene reported higher prevalence in men (Greene et al. 1976), while other researchers claimed that the formulation of fluorescein may be relevant to some reactions (Yannuzzi et al. 1986; Kwan et al. 2006). Probably the most intriguing factor in predicting SFA reactions is past SFA procedures and the personal history of reactions during previous SFAs. One major survey (Kwiterovich et al. 1991) concluded that patients undergoing their first SFA show an increased occurrence of ARs (5.1%) compared with those with previous uncomplicated SFAs (1.8%). As expected, the possibility of repeating an AR in future fluorescein administration is significantly higher (48.6%). In our study, 68% of the reactors also had a history of previous SFA reactions. The majority of severe SFA ARs are attributed to direct or immunologically mediated mast-cell and basophil degranulation (Lieberman et al. 2005). Most of these episodes are described as anaphylactoid based on the development of anaphylactic symptoms without previous sensitization or the presence of specific IgE against fluorescein. IgE-mediated anaphylactic reactions to intravenous fluorescein can be diagnosed by means of allergy skin testing (Trindade-Porto et al. 1999; Brockow et al. 2002). In two published cases of allergy to fluorescein documented with in vivo testing, patients were successfully desensitized and were able to undergo SFA without any reaction (Nucera et al. 2003; Knowles et al. 2007). It must be emphasized that desensitization protocols do not eliminate the danger of allergic reactions and should be applied only as an absolute necessity and before every single SFA. Several risk factors have been identified for increasing either the probability or the severity of anaphylactic and anaphylactoid reactions to contrast media and should also be evaluated in patients scheduled for SFA. These factors include atopic diseases (especially asthma) as well as previous drug reactions, which non-specifically increase the possibility of new ARs (Lieberman et al. 2005). A retrospective study did not show increased prevalence of ARs to SFAs in hypertensive and elderly patients (Musa et al. 2006). Bearing in mind that skin testing is not related to irritant reactions, we propose performing skin-prick tests on at-risk patients (including those with previous SFA reactions) using the undiluted preparation of sodium fluorescein in order to eliminate possible IgE sensitization. Reactors can take premedication consisting of oral corticosteroids, H1 and H2 antihistamines. Nevertheless, a prophylactic treatment is suggested by many authors even in the event of a negative skin test (Elis et al. 1980; Kwan et al. 2006). However, pretreatment does not eliminate the risk of anaphylaxis and resuscitation equipment should be available. For definite reactors to previous SFA with positive skin tests to fluorescein, future SFAs, when imperative, may be conducted according to specific desensitization protocols and/or under premedication – again, after receiving all measures necessary to treat possible anaphylaxis. For patients with a history of serious ARs to fluorescein administration, regardless of the skin-test results, future SFAs should be avoided unless they are considered absolutely necessary. In conclusion, this study expands in a larger patient population our previous finding that allergy skin testing cannot predict ARs to SFAs (Kalogeromitros et al. 2007). The inclusion of more patients was considered essential in consolidating these findings. Furthermore, we revealed a case of positive fluorescein skin test that was investigated further. Our study confirms previous reports that have shown limited value for allergy skin testing as a screening tool for serious ARs to fluorescein. The exact incidence of true IgE-mediated allergic reactions has not been estimated yet, but must be extremely low. Despite the fact that allergy skin tests to fluorescein are not standardized, they are the method of choice for diagnosing IgE-mediated hypersensitivity to fluorescein. The authors would like to thank Dr Eugenia Toumbis-Ioannou for her kind assistance in language revision.