- Email: [email protected]
Sensitivity and specificity of prick and intradermal testing in predicting response to nasal provocation with timothy grass antigen
Sensitivity and specificity of prick and intradermal testing in predicting response to nasal provocation with timothy grass antigen JOHN H. KROUSE,
MD, PHD,
KAMRAN SADRAZODI,
MD,
and KRISTY KERSWILL,
OBJECTIVE: Examine the efficacy of epicutaneous and intradermal testing in predicting response to nasal provocation with timothy antigen. STUDY DESIGN AND SETTING: Prospective study. Subjects were tested with Multi-Test II (MT). Subjects with negative wheals were tested with a 1:500 intradermal injection of timothy. They had baseline assessment of nasal cross-sectional area (CSA) using acoustic rhinometry and underwent nasal provocation with increasing timothy concentrations. CSA was assessed and nasal visual analog scale (VAS) completed with each concentration. RESULTS: Sensitivity and specificity of MT in predicting nasal response to provocation were 87% and 86%. Intradermal testing increased sensitivity to 93%. Hierarchical linear modeling (HLM) demonstrated that subjects positive to skin testing had significant reductions in CSA and worsening VAS scores with increasing concentrations of nasally delivered antigen. CONCLUSION AND SIGNIFICANCE: Epicutaneous testing with the Multi-Test II offers 87% sensitivity and 86% specificity in assessing timothy grass reactivity. (Otolaryngol Head Neck Surg 2004;131:215-9.)
I
ntradermal dilutional testing (IDT) has been used as a method for the testing of inhalant allergy for over 70 years. It is widely practiced by otolaryngic allergists, and can be traced to work done by Hansel in the 1920s.1 Observations by Phillips correlating the size of an intradermal wheal with the safe starting dose for immunotherapy further demonstrated the efficacy of this technique.2 One specific approach to IDT, skin endpoint titration (SET), became standardized in the 1940s and 1950s by Rinkel, who noted that a 5-fold dilution of antigenic extracts was clinically useful in guiding the management From the Department of Otolaryngology, Wayne State University. Presented at the Annual Meeting of the American Academy of Otolaryngic Allergy, Orlando, Florida, September 19, 2003. Study supported in part by a grant from Lincoln Diagnostics, Inc., Decatur, IL. Reprint requests: John H. Krouse, MD, PhD, Department of Otolaryngology, Wayne State University, 540 E. Canfield, 5E-UHC, Detroit, MI 48201; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2004.03.024
BSC,
Detroit, Michigan
of inhalant immunotherapy.3 A nasal challenge study by Krouse and Krouse that evaluated the efficacy of immunotherapy based on IDT demonstrated significant reduction in both symptomatology and nasal responsivity following 6 months of immunotherapy.4 In recent years, several commercial products have been developed that offer an alternative to IDT. One of these products is the Multi-Test II (Lincoln Diagnostics, Decatur, IL, USA), an epicutaneous testing device with 8 heads that deliver uniform punctures into the epidermis at a slightly deeper level than that obtained with standard prick testing. It has been demonstrated to outperform 4 other devices for skin prick testing, with a reported sensitivity of 100% and specificity of 95.5%.5 The absence of whealing with the Multi-Test II has been shown to agree well with the absence of specific IgE on serum RAST testing, while positive results have correlated less well.6 The results noted with testing from the Multi-Test II with various antigens in one study have been estimated to correlate with an SET dilution of between #3 (1:2500) and #4 (1: 12,500).7 One question that has been of concern for otolaryngic allergists is the meaning of positive intradermal tests at very low dilutions, such as #2 (1:500) and #1 (1:100). While some allergists will interpret the #1 dilution as positive and incorporate that antigen into the treatment vial, many others view this reaction as a nonspecific false positive. This variability is widely noted, and is based on differing interpretations of positivity and on anecdotal observations. In fact, in one recent study these positive intradermal tests were shown to be less specific than prick testing in representing true inhalant allergy.8 The assumption by those allergists who do implement immunotherapy based on intradermal test results is that prick testing is less sensitive to the detection of allergy and that intradermal testing is warranted to increase the sensitivity of the testing battery. The purpose of this research was to assess the sensitivity and specificity of intradermal testing and epicutaneous testing using the Multi-Test II device in demonstrating the presence of allergy to timothy grass antigen. In addition, the research also was performed to assess the true clinical significance of intradermal positivity in patients who are Multi-Test II negative for 215
216 KROUSE et al
allergy to timothy but who are positive to the antigen at intradermal IDT dilution #2. Patients were assessed with a progressive dilutional nasal provocation as a “gold standard” of response, correlating the direct effects of increasing concentrations of antigen on the end organ with skin responsiveness. It was the hypothesis of this proposed research that nasal challenge testing would correlate well with Multi-Test II results and that additional testing with 1:500 intradermal injections would add little clinical utility to the diagnosis of timothy allergy. METHOD Sample The sample consisted of 37 adults with a history of seasonal allergic rhinitis, collected consecutively from a public advertisement. Three groups were defined by individual response to skin testing: 1) 16 subjects who were positive to timothy by Multi-Test II with a wheal that was 3 mm larger than a negative control; 2) 1 subject who was negative to testing with Multi-Test II but positive to timothy at an IDT dilution of #2; and 3) 20 subjects who were negative to testing for timothy both by Multi-Test II and at IDT dilution #2. Inclusion criteria were: ● History of allergic rhinitis ● Adults, age 18 to 70 ● Ability to consent to participation in the study Exclusion criteria were: ● Use of an oral antihistamine within one week ● Pregnancy at the time of the study ● Concurrent use of leukotriene receptor antagonists ● Concurrent use of 5-OH-lipoxygenase inhibitors ● Use of nasal or systemic corticosteroids within 1 week ● History of endoscopic, intranasal, or external sinus surgery Setting The study was conducted in the offices and clinics of University Otolaryngology, P.C., the faculty practice associated with the Wayne State University Department of Otolaryngology–Head and Neck Surgery. All subjects were volunteers accrued through open advertisement in the university community. Procedure Individuals with a history suggestive of seasonal allergic rhinitis (SAR) were identified as prospective candidates for this study. They must have been symptomatic with SAR for a minimum of 2 years preceding the study. Those individuals who satisfied the diagnostic and inclusion/exclusion criteria were invited to participate. At this point, the nature of the study was
Otolaryngology– Head and Neck Surgery September 2004
explained to each individual, who was then asked to become a subject in the study. The consent procedure was completed, and each subject was asked to read and sign the consent form. The study and consent procedures were reviewed and approved by the Human Investigation Committee of Wayne State University. Subjects who agreed to participate were then included in the study. All subjects were tested for inhalant allergy using 2 different methods of skin testing. They were first tested for 24 inhalant antigens, including timothy, using the Multi-Test II as the primary testing methodology, along with appropriate controls.9 At this point, subjects who were positive to timothy antigen with a wheal size 3 mm greater than a negative control were included in Group 1, the timothy-positive group. Subjects who were negative to timothy were then tested with a #2 dilution (1:500 w:v) intradermally. If positive to the #2 dilution (7-mm intradermal wheal and a negative glycerin control) these patients were included in Group 2, the timothy-#2 group. Subjects negative to timothy with both Multi-Test II and intradermal testing at the #2 dilution were included in Group 3, the timothy-negative group. It was assumed for this study that subjects who tested positive using the Multi-Test II would also test positive on intradermal testing. At this time, each subject then proceeded to the nasal provocation portion of the study. Challenge solutions consisted of progressive dilutions of timothy extract in 0.9% saline and 0.4% phenol. These dilutions included IDT #3 (1:2500), IDT #2 (1:500), and IDT #1 (1:100). The volume was delivered into each nostril with a 2-second spray from a standard DeVilbiss spray bottle. Each subject first completed a visual analog scale (VAS), reflecting his/her global allergic symptomatology prior to antigen challenge, as well as a baseline nasal cross-sectional area (CSA) by acoustic rhinometry (AR) (Eccovision, Hood Laboratories, Pembroke, MA, USA). The subject then received a control challenge, consisting of 0.9% saline, 50% glycerin, and 0.4% phenol, in order to demonstrate lack of response to the diluent medium. At the conclusion of a 5-minute period, the subject again completed the VAS and AR. After these studies were completed, the subject waited 10 minutes for any effects to abate, and was challenged with the #3 (1:2500) dilution of timothy antigen. After a 5-minute period, the subject again completed the VAS and AR. If there was a 15% reduction in nasal CSA, the procedure was terminated. If a lesser change was noted, after 10 minutes a challenge was given with the #2 (1:500) dilution of timothy. The same progression was followed until the patient was positive at one dilution or negative at all dilutions.
Otolaryngology– Head and Neck Surgery Volume 131 Number 3
KROUSE et al 217
Table 1. Multi-Test II response and nasal provocation
Table 2. Intradermal (ID) response and nasal provocation
# of subjects who reached 15% reduction in nasal crosssectional area (CSA)
Multi-Test II® Total
Negative Positive
# of subjects who reached 15% reduction in nasal crosssectional area (CSA)
Negative
Positive
Total
19 3 22
2 13 15
21 16 37
At the conclusion of the study, all subjects who completed the nasal challenge received a payment of $100 for their participation. Outcome Measures Outcome measures for the study included 1) acoustic rhinometry (AR) of the minimal CSA of the nasal cavity, and 2) visual analog scale (VAS) for subjective assessment of nasal symptoms. Acoustic rhinometry was conducted using an Eccovision rhinometer (Hood Laboratories, Pembroke, MA, USA). Minimal CSA of each nasal cavity was assessed prior to challenge. The side of the nose with the greater CSA was chosen for subsequent nasal provocation challenge, and that CSA was used as the baseline measurement. A positive response to antigen challenge was defined as a reduction in CSA by 15% or greater from baseline in that nasal cavity. A positive response at one dilution resulted in termination of the test and was judged positive for all more concentrated dilutions. The visual analog scale was completed prior to challenge and after each nasal challenge. The scale consisted of a 10-cm line drawn on paper with marks at 1-cm intervals. Subjects were asked to globally assess their symptoms, including congestion, stuffiness, itching, sneezing, and rhinorrhea, and placed a single mark along the VAS to demonstrate their perceived symptomatology. RESULTS On the basis of skin testing 37 consecutive adult volunteers with symptoms of seasonal allergic rhinitis were classified into one of three groups: 1. Positive to Multi-Test II (n ⫽ 16) 2. Negative to Multi-Test II but positive to intradermal testing (n ⫽ 1) 3. Negative to both Multi-Test II and intradermal testing (n ⫽ 20) Results of nasal provocation as a function of group based on Multi-Test II positivity are presented in Table 1. These same results including intradermal positivity are presented in Table 2.
Positive with either MultiTest II or ID Total
Negative Positive
Negative
Positive
Total
19 3
1 14
20 17
22
15
37
Analysis of nasal provocation results among patients positive and negative to timothy grass antigen by MultiTest II demonstrates that the technique has a sensitivity of 87% (13/15) and a specificity of 86% (19/22). In subjects who were negative to Multi-Test II testing, intradermal tests with 1:500 timothy grass antigen were placed. Only 1 of these 21 subjects had a 7-mm wheal to timothy grass antigen after 10 minutes, demonstrating a positive response by the definition used in this study. The addition of intradermal positivity to the results of the Multi-Test II increased the sensitivity to 93% (14/15) without changing the specificity. Both Multi-Test II and intradermal testing demonstrated good sensitivity in this study. To compare the percentage of patients reaching the 15% change in nasal CSA across the patient groups a chi-square analysis using Fisher’s exact test was performed. To model the change in nasal CSA and the VAS across the 4 provocation concentrations hierarchical linear modeling (HLM) was used. Separate analyses were performed for each of the outcome measures. HLM was used since not all subjects had complete data for all 4 possible concentration values. To compare the 2 patient groupings in the HLM analysis dummy coding was used. Chi-square analysis revealed a significant difference in the percentage of subjects reaching the 15% reduction in nasal CSA between those subjects positive to skin testing with Multi-Test II and those subjects negative to testing (2 ⫽ 19.38, df ⫽ 1, P ⬍ 0.001). In addition, using HLM analysis, subjects who were positive to skin testing demonstrated a significant reduction in nasal CSA as the concentration of nasal antigen increased when compared with those subjects who were negative to skin testing (t ⫽ ⫺3.192, df ⫽ 35, P ⬍ 0.001). Analysis of the VAS data using HLM demonstrated a similar difference in subjective nasal symptoms between those subjects testing positive and negative to timothy grass antigen by Multi-Test II. Subjects who
218 KROUSE et al
Otolaryngology– Head and Neck Surgery September 2004
Fig 1. Nasal cross-sectional area by acoustic rhinometry (AR) and concentration of nasally applied timothy grass antigen.
Fig 2. Visual analog scale (VAS) scores and concentration of nasally applied timothy grass antigen.
were positive to skin testing demonstrated more severe nasal symptoms with increasing concentrations of nasally delivered timothy antigen than those subjects who were negative to skin testing (t ⫽ 2.987, df ⫽ 35, P ⬍ 0.007).
DISCUSSION The results of the present study demonstrate that the Multi-Test II is an effective device for the detection of sensitivity to timothy grass. In fact, the sensitivity of this epicutaneous test is 87% in the present sample
Otolaryngology– Head and Neck Surgery Volume 131 Number 3
when evaluated against nasal provocation with timothy antigen. The Multi-Test II also demonstrates good specificity of 86% in the present sample. These clinimetric properties suggest that use of this technique can be of utility in testing for grass allergy. The current results also demonstrate that the use of a single intradermal test at a dilution of 1:500 for timothy grass adds little to the sensitivity of skin testing with the Multi-Test II. The sensitivity with the combined prick/intradermal approach for timothy grass is 93%, adding only about 6% to the sensitivity of the Multi-Test II alone. In fact, in only 1 of 21 negative prick tests was the intradermal testing with 1:500 timothy antigen positive. In the other 20 subjects, a negative Multi-Test II predicted a negative 1:500 intradermal test with timothy antigen. Analysis of the clinical indicators also demonstrated that sequential nasal provocation with timothy grass antigen has good validity as a method for assessing degree of responsiveness in timothy allergic patients. HLM analysis of slopes in the positive and negative skin test responders confirms that nasal provocation effectively differentiates these 2 groups (Fig 1). In addition, increasing concentrations of nasally delivered antigen are associated with increasing nasal symptoms as assessed by the VAS (Fig 2). These 2 findings support the use of nasal provocation testing as a valid “gold standard” for assessment of skin testing sensitivity and specificity. These current findings support the use of prick testing with the Multi-Test II as an accurate methodology for the diagnosis of timothy grass sensitivity. The addition of intradermal testing with highly concentrated antigens in the setting of negative prick testing offers little incremental sensitivity in the detection of timothy grass allergy. With the additional time and expense involved in performing intradermal tests, their routine use in assessing timothy grass sensitivity when prick testing is negative would appear to be unwarranted. In those cases in which a clinical history is strongly suggestive of timothy grass allergy and the Multi-Test II is negative, the use of an intradermal test in that select circumstance might be indicated. While the clinimetric properties of the Multi-Test II appear very good for testing with pollens such as timothy grass, these characteristics cannot be automatically applied to other antigens. It would appear that other allergenic proteins may not behave in a similar manner. For example, in a study by Krouse and colleagues,10 the sensitivity and specificity of the MultiTest II device for Alternaria when judged against inhaled nasal challenge of Alternaria antigen by acoustic rhinometry were only about 50% each. These findings suggest that skin testing for mold allergy may be more
KROUSE et al
219
complicated and less reliable than with antigens such as pollens. Additional research is necessary to further characterize differential responses to skin testing when using other antigens. It must finally be remembered that the Multi-Test II is a qualitative test of the presence or absence of allergic responsiveness on the skin. It is not an indicator of the precise degree of sensitivity, and the use of its wheal sizes for individual antigens does not accurately predict allergic endpoints.11 Prior to administration of immunotherapy, it is therefore recommended that either quantitative in vitro allergy testing or intradermal dilutional testing is used to more precisely estimate the allergic endpoint and prepare the indicated treatment vials. CONCLUSION Skin prick testing with the Multi-Test II demonstrates excellent sensitivity and specificity in its ability to detect the skin’s response to epicutaneous delivery of timothy grass antigen. In the presence of a negative whealing response, additional testing with a concentrated intradermal injection increased the sensitivity of the testing by only 6% in this study. Further research is necessary to evaluate the differential responses of other seasonal and perennial antigens to skin testing and to determine their individual clinimetric properties against end-organ provocation (Fig 1 and 2). REFERENCES 1. Keenan JP. History of skin testing and evolution of skin endpoint titration. In: Mabry RL, editor. Skin endpoint titration. New York: Thieme; 1992. p. 12-6. 2. Phillips EW. Relief of hayfever by intradermal injections of pollen extract. JAMA 1926;86:182-4. 3. Rinkel HJ. The management of clinical allergy. Part II, etiologic factors and skin titration. Arch Otolaryngol 1963;77:42-75. 4. Krouse JH, Krouse HJ. Efficacy of immunotherapy based on skin end-point titration. Otolaryngol Head Neck Surg 2000;123:183-7. 5. Nelson HS, Lahr J, Buchmeier A, McCormick D. Evaluation of devices for skin prick testing. J Allergy Clin Immunol 1998;101: 153-6. 6. Levine JL, Mabry RL, Mabry CL. Comparison of Multi-Test device skin testing and modified RAST results. Otolaryngol Head Neck Surg 1998;118:797-9. 7. Murphree JT, Kniker WT. Correlation of immediate skin test responses to antigens introduced by multi-test and intracutaneous routes. Ann Allergy 1979;43:279-85. 8. Gungor A, Houser SM, Aquino BF, et al. Comparing skin endpoint titration to other forms of allergy testing. ENT J 2004; 83:54-60. 9. Hurst DS, Gordon BR, Krouse JH. The importance of glycerincontaining controls in the interpretation of allergy skin tests. Otolaryngol Head Neck Surg 2002;127:177-81. 10. Krouse JH, Shah AG, Kerswill K. Skin testing in predicting response to nasal provocation with Alternaria. Laryngoscope, in press. 11. Coutras SW. Can the Multi-Test II predict the endpoint of the intradermal dilution test? Paper presented at the Annual Meeting of the American Academy of Otolaryngic Allergy, September 19, 2003.
MD, PHD,
KAMRAN SADRAZODI,
MD,
and KRISTY KERSWILL,
OBJECTIVE: Examine the efficacy of epicutaneous and intradermal testing in predicting response to nasal provocation with timothy antigen. STUDY DESIGN AND SETTING: Prospective study. Subjects were tested with Multi-Test II (MT). Subjects with negative wheals were tested with a 1:500 intradermal injection of timothy. They had baseline assessment of nasal cross-sectional area (CSA) using acoustic rhinometry and underwent nasal provocation with increasing timothy concentrations. CSA was assessed and nasal visual analog scale (VAS) completed with each concentration. RESULTS: Sensitivity and specificity of MT in predicting nasal response to provocation were 87% and 86%. Intradermal testing increased sensitivity to 93%. Hierarchical linear modeling (HLM) demonstrated that subjects positive to skin testing had significant reductions in CSA and worsening VAS scores with increasing concentrations of nasally delivered antigen. CONCLUSION AND SIGNIFICANCE: Epicutaneous testing with the Multi-Test II offers 87% sensitivity and 86% specificity in assessing timothy grass reactivity. (Otolaryngol Head Neck Surg 2004;131:215-9.)
I
ntradermal dilutional testing (IDT) has been used as a method for the testing of inhalant allergy for over 70 years. It is widely practiced by otolaryngic allergists, and can be traced to work done by Hansel in the 1920s.1 Observations by Phillips correlating the size of an intradermal wheal with the safe starting dose for immunotherapy further demonstrated the efficacy of this technique.2 One specific approach to IDT, skin endpoint titration (SET), became standardized in the 1940s and 1950s by Rinkel, who noted that a 5-fold dilution of antigenic extracts was clinically useful in guiding the management From the Department of Otolaryngology, Wayne State University. Presented at the Annual Meeting of the American Academy of Otolaryngic Allergy, Orlando, Florida, September 19, 2003. Study supported in part by a grant from Lincoln Diagnostics, Inc., Decatur, IL. Reprint requests: John H. Krouse, MD, PhD, Department of Otolaryngology, Wayne State University, 540 E. Canfield, 5E-UHC, Detroit, MI 48201; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2004.03.024
BSC,
Detroit, Michigan
of inhalant immunotherapy.3 A nasal challenge study by Krouse and Krouse that evaluated the efficacy of immunotherapy based on IDT demonstrated significant reduction in both symptomatology and nasal responsivity following 6 months of immunotherapy.4 In recent years, several commercial products have been developed that offer an alternative to IDT. One of these products is the Multi-Test II (Lincoln Diagnostics, Decatur, IL, USA), an epicutaneous testing device with 8 heads that deliver uniform punctures into the epidermis at a slightly deeper level than that obtained with standard prick testing. It has been demonstrated to outperform 4 other devices for skin prick testing, with a reported sensitivity of 100% and specificity of 95.5%.5 The absence of whealing with the Multi-Test II has been shown to agree well with the absence of specific IgE on serum RAST testing, while positive results have correlated less well.6 The results noted with testing from the Multi-Test II with various antigens in one study have been estimated to correlate with an SET dilution of between #3 (1:2500) and #4 (1: 12,500).7 One question that has been of concern for otolaryngic allergists is the meaning of positive intradermal tests at very low dilutions, such as #2 (1:500) and #1 (1:100). While some allergists will interpret the #1 dilution as positive and incorporate that antigen into the treatment vial, many others view this reaction as a nonspecific false positive. This variability is widely noted, and is based on differing interpretations of positivity and on anecdotal observations. In fact, in one recent study these positive intradermal tests were shown to be less specific than prick testing in representing true inhalant allergy.8 The assumption by those allergists who do implement immunotherapy based on intradermal test results is that prick testing is less sensitive to the detection of allergy and that intradermal testing is warranted to increase the sensitivity of the testing battery. The purpose of this research was to assess the sensitivity and specificity of intradermal testing and epicutaneous testing using the Multi-Test II device in demonstrating the presence of allergy to timothy grass antigen. In addition, the research also was performed to assess the true clinical significance of intradermal positivity in patients who are Multi-Test II negative for 215
216 KROUSE et al
allergy to timothy but who are positive to the antigen at intradermal IDT dilution #2. Patients were assessed with a progressive dilutional nasal provocation as a “gold standard” of response, correlating the direct effects of increasing concentrations of antigen on the end organ with skin responsiveness. It was the hypothesis of this proposed research that nasal challenge testing would correlate well with Multi-Test II results and that additional testing with 1:500 intradermal injections would add little clinical utility to the diagnosis of timothy allergy. METHOD Sample The sample consisted of 37 adults with a history of seasonal allergic rhinitis, collected consecutively from a public advertisement. Three groups were defined by individual response to skin testing: 1) 16 subjects who were positive to timothy by Multi-Test II with a wheal that was 3 mm larger than a negative control; 2) 1 subject who was negative to testing with Multi-Test II but positive to timothy at an IDT dilution of #2; and 3) 20 subjects who were negative to testing for timothy both by Multi-Test II and at IDT dilution #2. Inclusion criteria were: ● History of allergic rhinitis ● Adults, age 18 to 70 ● Ability to consent to participation in the study Exclusion criteria were: ● Use of an oral antihistamine within one week ● Pregnancy at the time of the study ● Concurrent use of leukotriene receptor antagonists ● Concurrent use of 5-OH-lipoxygenase inhibitors ● Use of nasal or systemic corticosteroids within 1 week ● History of endoscopic, intranasal, or external sinus surgery Setting The study was conducted in the offices and clinics of University Otolaryngology, P.C., the faculty practice associated with the Wayne State University Department of Otolaryngology–Head and Neck Surgery. All subjects were volunteers accrued through open advertisement in the university community. Procedure Individuals with a history suggestive of seasonal allergic rhinitis (SAR) were identified as prospective candidates for this study. They must have been symptomatic with SAR for a minimum of 2 years preceding the study. Those individuals who satisfied the diagnostic and inclusion/exclusion criteria were invited to participate. At this point, the nature of the study was
Otolaryngology– Head and Neck Surgery September 2004
explained to each individual, who was then asked to become a subject in the study. The consent procedure was completed, and each subject was asked to read and sign the consent form. The study and consent procedures were reviewed and approved by the Human Investigation Committee of Wayne State University. Subjects who agreed to participate were then included in the study. All subjects were tested for inhalant allergy using 2 different methods of skin testing. They were first tested for 24 inhalant antigens, including timothy, using the Multi-Test II as the primary testing methodology, along with appropriate controls.9 At this point, subjects who were positive to timothy antigen with a wheal size 3 mm greater than a negative control were included in Group 1, the timothy-positive group. Subjects who were negative to timothy were then tested with a #2 dilution (1:500 w:v) intradermally. If positive to the #2 dilution (7-mm intradermal wheal and a negative glycerin control) these patients were included in Group 2, the timothy-#2 group. Subjects negative to timothy with both Multi-Test II and intradermal testing at the #2 dilution were included in Group 3, the timothy-negative group. It was assumed for this study that subjects who tested positive using the Multi-Test II would also test positive on intradermal testing. At this time, each subject then proceeded to the nasal provocation portion of the study. Challenge solutions consisted of progressive dilutions of timothy extract in 0.9% saline and 0.4% phenol. These dilutions included IDT #3 (1:2500), IDT #2 (1:500), and IDT #1 (1:100). The volume was delivered into each nostril with a 2-second spray from a standard DeVilbiss spray bottle. Each subject first completed a visual analog scale (VAS), reflecting his/her global allergic symptomatology prior to antigen challenge, as well as a baseline nasal cross-sectional area (CSA) by acoustic rhinometry (AR) (Eccovision, Hood Laboratories, Pembroke, MA, USA). The subject then received a control challenge, consisting of 0.9% saline, 50% glycerin, and 0.4% phenol, in order to demonstrate lack of response to the diluent medium. At the conclusion of a 5-minute period, the subject again completed the VAS and AR. After these studies were completed, the subject waited 10 minutes for any effects to abate, and was challenged with the #3 (1:2500) dilution of timothy antigen. After a 5-minute period, the subject again completed the VAS and AR. If there was a 15% reduction in nasal CSA, the procedure was terminated. If a lesser change was noted, after 10 minutes a challenge was given with the #2 (1:500) dilution of timothy. The same progression was followed until the patient was positive at one dilution or negative at all dilutions.
Otolaryngology– Head and Neck Surgery Volume 131 Number 3
KROUSE et al 217
Table 1. Multi-Test II response and nasal provocation
Table 2. Intradermal (ID) response and nasal provocation
# of subjects who reached 15% reduction in nasal crosssectional area (CSA)
Multi-Test II® Total
Negative Positive
# of subjects who reached 15% reduction in nasal crosssectional area (CSA)
Negative
Positive
Total
19 3 22
2 13 15
21 16 37
At the conclusion of the study, all subjects who completed the nasal challenge received a payment of $100 for their participation. Outcome Measures Outcome measures for the study included 1) acoustic rhinometry (AR) of the minimal CSA of the nasal cavity, and 2) visual analog scale (VAS) for subjective assessment of nasal symptoms. Acoustic rhinometry was conducted using an Eccovision rhinometer (Hood Laboratories, Pembroke, MA, USA). Minimal CSA of each nasal cavity was assessed prior to challenge. The side of the nose with the greater CSA was chosen for subsequent nasal provocation challenge, and that CSA was used as the baseline measurement. A positive response to antigen challenge was defined as a reduction in CSA by 15% or greater from baseline in that nasal cavity. A positive response at one dilution resulted in termination of the test and was judged positive for all more concentrated dilutions. The visual analog scale was completed prior to challenge and after each nasal challenge. The scale consisted of a 10-cm line drawn on paper with marks at 1-cm intervals. Subjects were asked to globally assess their symptoms, including congestion, stuffiness, itching, sneezing, and rhinorrhea, and placed a single mark along the VAS to demonstrate their perceived symptomatology. RESULTS On the basis of skin testing 37 consecutive adult volunteers with symptoms of seasonal allergic rhinitis were classified into one of three groups: 1. Positive to Multi-Test II (n ⫽ 16) 2. Negative to Multi-Test II but positive to intradermal testing (n ⫽ 1) 3. Negative to both Multi-Test II and intradermal testing (n ⫽ 20) Results of nasal provocation as a function of group based on Multi-Test II positivity are presented in Table 1. These same results including intradermal positivity are presented in Table 2.
Positive with either MultiTest II or ID Total
Negative Positive
Negative
Positive
Total
19 3
1 14
20 17
22
15
37
Analysis of nasal provocation results among patients positive and negative to timothy grass antigen by MultiTest II demonstrates that the technique has a sensitivity of 87% (13/15) and a specificity of 86% (19/22). In subjects who were negative to Multi-Test II testing, intradermal tests with 1:500 timothy grass antigen were placed. Only 1 of these 21 subjects had a 7-mm wheal to timothy grass antigen after 10 minutes, demonstrating a positive response by the definition used in this study. The addition of intradermal positivity to the results of the Multi-Test II increased the sensitivity to 93% (14/15) without changing the specificity. Both Multi-Test II and intradermal testing demonstrated good sensitivity in this study. To compare the percentage of patients reaching the 15% change in nasal CSA across the patient groups a chi-square analysis using Fisher’s exact test was performed. To model the change in nasal CSA and the VAS across the 4 provocation concentrations hierarchical linear modeling (HLM) was used. Separate analyses were performed for each of the outcome measures. HLM was used since not all subjects had complete data for all 4 possible concentration values. To compare the 2 patient groupings in the HLM analysis dummy coding was used. Chi-square analysis revealed a significant difference in the percentage of subjects reaching the 15% reduction in nasal CSA between those subjects positive to skin testing with Multi-Test II and those subjects negative to testing (2 ⫽ 19.38, df ⫽ 1, P ⬍ 0.001). In addition, using HLM analysis, subjects who were positive to skin testing demonstrated a significant reduction in nasal CSA as the concentration of nasal antigen increased when compared with those subjects who were negative to skin testing (t ⫽ ⫺3.192, df ⫽ 35, P ⬍ 0.001). Analysis of the VAS data using HLM demonstrated a similar difference in subjective nasal symptoms between those subjects testing positive and negative to timothy grass antigen by Multi-Test II. Subjects who
218 KROUSE et al
Otolaryngology– Head and Neck Surgery September 2004
Fig 1. Nasal cross-sectional area by acoustic rhinometry (AR) and concentration of nasally applied timothy grass antigen.
Fig 2. Visual analog scale (VAS) scores and concentration of nasally applied timothy grass antigen.
were positive to skin testing demonstrated more severe nasal symptoms with increasing concentrations of nasally delivered timothy antigen than those subjects who were negative to skin testing (t ⫽ 2.987, df ⫽ 35, P ⬍ 0.007).
DISCUSSION The results of the present study demonstrate that the Multi-Test II is an effective device for the detection of sensitivity to timothy grass. In fact, the sensitivity of this epicutaneous test is 87% in the present sample
Otolaryngology– Head and Neck Surgery Volume 131 Number 3
when evaluated against nasal provocation with timothy antigen. The Multi-Test II also demonstrates good specificity of 86% in the present sample. These clinimetric properties suggest that use of this technique can be of utility in testing for grass allergy. The current results also demonstrate that the use of a single intradermal test at a dilution of 1:500 for timothy grass adds little to the sensitivity of skin testing with the Multi-Test II. The sensitivity with the combined prick/intradermal approach for timothy grass is 93%, adding only about 6% to the sensitivity of the Multi-Test II alone. In fact, in only 1 of 21 negative prick tests was the intradermal testing with 1:500 timothy antigen positive. In the other 20 subjects, a negative Multi-Test II predicted a negative 1:500 intradermal test with timothy antigen. Analysis of the clinical indicators also demonstrated that sequential nasal provocation with timothy grass antigen has good validity as a method for assessing degree of responsiveness in timothy allergic patients. HLM analysis of slopes in the positive and negative skin test responders confirms that nasal provocation effectively differentiates these 2 groups (Fig 1). In addition, increasing concentrations of nasally delivered antigen are associated with increasing nasal symptoms as assessed by the VAS (Fig 2). These 2 findings support the use of nasal provocation testing as a valid “gold standard” for assessment of skin testing sensitivity and specificity. These current findings support the use of prick testing with the Multi-Test II as an accurate methodology for the diagnosis of timothy grass sensitivity. The addition of intradermal testing with highly concentrated antigens in the setting of negative prick testing offers little incremental sensitivity in the detection of timothy grass allergy. With the additional time and expense involved in performing intradermal tests, their routine use in assessing timothy grass sensitivity when prick testing is negative would appear to be unwarranted. In those cases in which a clinical history is strongly suggestive of timothy grass allergy and the Multi-Test II is negative, the use of an intradermal test in that select circumstance might be indicated. While the clinimetric properties of the Multi-Test II appear very good for testing with pollens such as timothy grass, these characteristics cannot be automatically applied to other antigens. It would appear that other allergenic proteins may not behave in a similar manner. For example, in a study by Krouse and colleagues,10 the sensitivity and specificity of the MultiTest II device for Alternaria when judged against inhaled nasal challenge of Alternaria antigen by acoustic rhinometry were only about 50% each. These findings suggest that skin testing for mold allergy may be more
KROUSE et al
219
complicated and less reliable than with antigens such as pollens. Additional research is necessary to further characterize differential responses to skin testing when using other antigens. It must finally be remembered that the Multi-Test II is a qualitative test of the presence or absence of allergic responsiveness on the skin. It is not an indicator of the precise degree of sensitivity, and the use of its wheal sizes for individual antigens does not accurately predict allergic endpoints.11 Prior to administration of immunotherapy, it is therefore recommended that either quantitative in vitro allergy testing or intradermal dilutional testing is used to more precisely estimate the allergic endpoint and prepare the indicated treatment vials. CONCLUSION Skin prick testing with the Multi-Test II demonstrates excellent sensitivity and specificity in its ability to detect the skin’s response to epicutaneous delivery of timothy grass antigen. In the presence of a negative whealing response, additional testing with a concentrated intradermal injection increased the sensitivity of the testing by only 6% in this study. Further research is necessary to evaluate the differential responses of other seasonal and perennial antigens to skin testing and to determine their individual clinimetric properties against end-organ provocation (Fig 1 and 2). REFERENCES 1. Keenan JP. History of skin testing and evolution of skin endpoint titration. In: Mabry RL, editor. Skin endpoint titration. New York: Thieme; 1992. p. 12-6. 2. Phillips EW. Relief of hayfever by intradermal injections of pollen extract. JAMA 1926;86:182-4. 3. Rinkel HJ. The management of clinical allergy. Part II, etiologic factors and skin titration. Arch Otolaryngol 1963;77:42-75. 4. Krouse JH, Krouse HJ. Efficacy of immunotherapy based on skin end-point titration. Otolaryngol Head Neck Surg 2000;123:183-7. 5. Nelson HS, Lahr J, Buchmeier A, McCormick D. Evaluation of devices for skin prick testing. J Allergy Clin Immunol 1998;101: 153-6. 6. Levine JL, Mabry RL, Mabry CL. Comparison of Multi-Test device skin testing and modified RAST results. Otolaryngol Head Neck Surg 1998;118:797-9. 7. Murphree JT, Kniker WT. Correlation of immediate skin test responses to antigens introduced by multi-test and intracutaneous routes. Ann Allergy 1979;43:279-85. 8. Gungor A, Houser SM, Aquino BF, et al. Comparing skin endpoint titration to other forms of allergy testing. ENT J 2004; 83:54-60. 9. Hurst DS, Gordon BR, Krouse JH. The importance of glycerincontaining controls in the interpretation of allergy skin tests. Otolaryngol Head Neck Surg 2002;127:177-81. 10. Krouse JH, Shah AG, Kerswill K. Skin testing in predicting response to nasal provocation with Alternaria. Laryngoscope, in press. 11. Coutras SW. Can the Multi-Test II predict the endpoint of the intradermal dilution test? Paper presented at the Annual Meeting of the American Academy of Otolaryngic Allergy, September 19, 2003.