Cow’s milk allergy skin tests: fresh milk, commercial extracts, or both?

From 2010 through 2018, 698 patients with a diagnosis of "CMA", were evaluated. Among them, 272 (39.0%) were excluded based on the exclusion criteria, and another 38 were lost to follow-up (shown in Fig. 1). Of the 388 patients in the study cohort, 215 (55.4%) were male.

Fig. 1figure 1

Patient flow diagram. CMA cow's milk allergy; OFC oral food challenge; SPT skin prick test

The study cohort of 388 patients included the following subgroups: Group A (allergic) 227(58.5%), of whom 30(7.7%) were proven by OFC (group A1), and 197(50.8%) by SPT (group A2). Group B (non-allergic) included 161 subjects (41.5%) who did not react during the OFC.

There were no significant differences between the three groups regarding age at the first allergic reaction (shown in Table 1). Additional demographic parameters are presented in Table 1.

Table 1 Comparison of demographic and clinical parameters between the 3 study groupsAtopic comorbidities

Overall, 134 patients (34.5%) had at least one additional atopic disease; 82(21.1%) had asthma and 68 (17.5%) had atopic dermatitis. Food allergies other than milk were found in 78 patients (20.1%). The prevalence of asthma was similar between group A1 and A2 (p = NS). The incidence in groups A1 and A2 compared to group B was statistically significant (22.58% for group A1 and 31.12% for group A2 vs 8.75% for group B; p < 0.05).

The incidence of other food allergy was significantly more common in group A2 when compared to group B (25.5% vs 14.4%, respectively p < 0.05) (shown in Table 1).

Symptoms during the first reaction

Urticarial rash was the most common symptom affecting 335 (86.3%) patients, 55 (14.5%) had breathing difficulties and 212 (55.9%) had vomiting. Breathing difficulties were more prevalent in group A2 as compared to group B (22.9% vs 4.5%, p < 0.001; shown in Table 1). No other symptoms were reported.

Skin tests

All subjects (n = 388) had done SPT with milk extract and casein simultaneously. Of them, 322 (83%) subjects had done also SPT with FM. From the entire cohort, 66 (17%) were examined only once and have results for the first SPT alone. From those who had last SPT’s, 274 (85%) had done also SPT with FM (shown in Additional file 1: Table S1).

Skin test with FM

The wheal size with FM was significantly larger than with milk extract and casein at the first SPT recorded and in the last SPT recorded in the total cohort and for each group separately: at the first SPT—5.1 mm, CI 95%(4.62–5.6) vs. 5.91 mm, CI 95%(5.53–6.3) vs. 8.88 mm, CI 95%(8.24–9.52), p < 0.05, for casein, milk extract and FM respectively. At the last SPT—3.74 mm, CI 95%(3.26–4.21) vs. 5.02 mm, CI95%(4.56–5.48) vs. 8.21 mm, CI 95%(7.52–8.9), for casein, milk extract and FM respectively; p < 0.05 (shown in Fig. 2).

Fig. 2figure 2

Skin prick tests wheal size (mm) in the different study groups. A Mean skin prick test results at diagnosis. B Mean last skin prick test results. Only significant differences are shown

First SPT

The time elapsed between the first recorded reaction and the first SPT was 11.51 ± 0.72 months.

The mean SPT results of the first tests are shown in Fig. 2A. Significant differences between the wheal size in the different groups were found only in SPT with casein (5.27 mm, CI 95%(3.02–7.52), 6.68 mm, CI 95%(5.99–7.37), 3.12 mm, CI 95%(2.53–3.7), p < 0.05 for groups A1, A2, and B, respectively). Significant differences in the wheal size of SPT with milk extract were found only between group B and group A2 (4.67 mm, CI 95%(4.19–5.15) vs. 6.95 mm, CI 95%(6.39–7.51), respectively). There were no significant differences between the wheal size in the first SPT with FM among the groups (8.43 mm, CI 95%(6.71–10.15), 9.63 mm, CI 95%(8.82–10.44), 8.04 mm, CI 95%(6.9–9.18) for groups A1, A2, and B, respectively).

Only two patients in the not allergic group (1.25%) had first casein SPT ≥ 14 mm.

A binary adjusted logistic regression models have shown significant association between the wheal size of the first SPT with casein and the OFC results (OR = 1.148, CI 95%(1.046–1.259), p < 0.05; shown in Additional file 1: Table S2A). These results are in line with the unadjusted results.

Last SPT

The time elapsed between the first recorded SPT and the last was 2.48 ± 0.13 years.

The mean SPT results of the last tests recorded are shown in Fig. 2B. For patients who underwent OFC, the last SPT results are those prior to the OFC. Significant differences between the wheal sizes in the different groups were found in SPT with all the extracts, including FM.

Logistic regression models after adjustment showed that the association between the wheal size of the last SPT with each extract and the OFC results was statistically significant (shown in Additional file 1: Table S2B). However, the OR for casein was significantly higher than the OR of milk extract and FM (1.907, 1.354 and 1.151, respectively).

Ratio between the first SPT and the last SPT

The ratio \(\frac\) was < 1 for casein, milk extract, and FM in group B, and casein in group A1. The ratio was > 1 for all extracts in group A2 and milk extract and FM in group A1 (shown in Fig. 3). The ratio was significantly lower in group B compared to the two other groups for all three types of SPT extracts (p < 0.05), except for FM where the difference was significant between groups B and A2, and not between groups B and A1. There were no significant differences between the two allergic groups. Logistic regression models after adjustment showed that each extract ratio was statistically significant. The maximal OR was 2.922 (CI 95% (1.317–6.481), p < 0.01) for casein SPT ratio, but without significant differences for SPT ratio with milk extract and FM (shown in Additional file 1: Table S2C).

Fig. 3figure 3

The ratio between the last skin prick test and the first skin prick test recorded according to the study groups. OFC oral food challenge; SPT skin prick test. The valid number of SPT for calculations—Not allergic (casein—145, milk extract—147, fresh milk—118), allergic by OFC (casein—27, milk extract—28, fresh milk—25), and allergic by SPT (casein—140, milk extract—143, fresh milk—129)

Cutoff point for the different extracts

Standardization of wheal size to age led to the equations:

$$}\left( }} \right) \, = . + \left( *}} \right),$$

$$}\left( }} \right) = .0 + \left( *}} \right).$$

For casein, the correlation between age and wheal size was not significant, and wheal size < 3 mm was defined as negative for all ages.

From this point, negative and positive SPT results were defined accordingly.

The NPV of all extracts separately or in combination ranged 92.76–94.2%, without significant differences. The PPV was maximal with the combination of all extracts with 72.73%, the sensitivity was maximal with FM 84%, and the specificity was maximal with the combination of all three extracts with a rate of 95.92% (range 55.08–95.92%). The OR was maximal for the combination of all three extracts and minimal with FM (34.18, 6.44 respectively) (shown in Table 2). NPV, PPV, sensitivity, and specificity were calculated only for groups A1 and B.

Table 2 Predictive values of the different extracts

The AUC for SPT ratio of milk extract and FM were under the accepted value for discrimination (0.67 and 0.68 respectively). The AUC for SPT ratio of casein was 0.73. ROC curve analyses for the ratio found that the optimal sensitivity and specificity for the SPT ratio with casein was 0.68 (OR = 5.36, p < 0.001; shown in Fig. 4). There was no optimal value for milk extract and FM. The ROC curve was created for groups A1 and B, excluding group A2 that was defined as allergic without OFC.

Fig. 4figure 4

ROC curve for the ratio between the last and the first SPT with the different extracts. ROC curve created for groups A1 and B (Allergic by oral food challenge and Not allergic by oral food challenge)

留言 (0)

沒有登入
gif