HTML
-
A total of 1903 reports were picked in our study for screening after removing 2447 duplicates by endnote software. Subsequently, 255 papers were eligible for full text screening against our inclusion and exclusion criteria which resulted in inclusion of 50 reports. An additional 7 papers were included as a result of the manual search. Finally, 57 studies were included in our study (Fig. 1). Upon the basis of study design; studies were divided as the following, 23 cohort, 17 crosssectional, 13 case-control, and 4 case series studies.In terms of risk of bias, all studies obtained fair criteria. The eligible included studies were grouped according to the reported allergy-related symptoms and markers into pruritus, asthma, non-specified allergy (in which allergy is not specified in included papers), food allergy, skin allergy, urticaria, IgE positivity, total IgE concentration, and specific IgE concentration (Supplementary Table S1 and S2).
-
Pruritus was the most common allergic sign in dengue patients irrespective of dengue severity with a percent of 28.6%. Subsequently; followed by non-specified allergy and asthma with the percent of 13% and 6.5%, respectively. Only one study reported urticaria as an allergic symptom of dengue patients revealed a rate of 3.7% (Fig. 2). No publication bias was found for pruritus and asthma (P = 0.7) and (P = 0.9), respectively (Supplementary Figures S1, S2).
-
Four studies compared pruritus among dengue patients and non-dengue patients (Deparis et al. 1998; Vasconcelos et al. 1998; Passos et al. 2008; Chen et al. 2017). Significance was found between the two groups (OR 2.79, 95% CI 2.08–3.73, P < 0.001) without heterogeneity (Table 1). Total IgE level was significantly higher in dengue compared to healthy control group (SMD 0.78, 95% CI 0.22–1.34, P = 0.006) (Fig. 3) (Pavri et al. 1977; 1979; Mabalirajan et al. 2005); however the significance was lost when comparing DHF and healthy controls (P>0.05) (Pavri et al. 1977, 1979).
Variable Group 1 Group 2 No. of studies Heterogeneity Model Overall effect Name Number/total Name Number/total P value I2 (%) P value Effect size a (95%CI) Asthma DHF 97/1686 DF 210/4359 5 0.26 23.5 Fixed 0.18 1.19 (0.92–1.54) Asthmac DHF 61/868 DF 136/2892 4 0.62 0 Fixed 0.02 1.47 (1.06–2.04) Pruritis Dengue 476/1026 Non dengue 86/385 4 0.62 0 Fixed 0 2.79 (2.08–3.73) Pruritis DHF 262/544 DF 13, 353/48, 452 2 0.02 80.7 Random 0.1 2.06 (0.87 to - 4.89) Total IgE Dengue 72 Controls 101 3 0.06 64 Random 0.006 0.78 (0.22–1.34)b Total IgE DHF 44 Controls 41 2 0.03 78 Random 0.15 0.74 (- 0.28 to 1.75)b Dengue specific Seconda 148 Primary 100 3 0.88 0 Fixed 0 1.50 (1.52–1.78)b IgE level DF DF Dengue specific Seconda 148 Secondary 132 3 0.02 74.4 Random 0.95 0.02 (- 0.48 to 0.51)b IgE level DF DHF IgE positive DHF 45/53 DF 40/79 3 0.66 0 Fixed 0.01 4.44 (1.43–13.77) Allergy (non specified) DHF 174/660 DF 1127/2491 2 0 99 Random 0.53 0.49 (0.05–4.57) Food allergy DHF 30/490 DF 51/1316 1d 0.47 0 Fixed 0.05 1.59 (1.00–2.53) Skin allergy DHF 36/490 DF 65/1316 1d 0.24 26.6 Fixed 0.05 1.53 (1.00–2.34) aOdds Ratio otherwise specifically indicated.
bSDM standard difference in means.
cSensitivity analysis after removal of Pang/2012/Singapore.
dTwo data sets from the same study were pooled.Table 1. Results of the meta-analysis testing the association between dengue and allergic signs.
-
the Allergic Symptoms and Markers The pooled odds ratio obtained from studies comparing asthma between DHF and DF illustrated no significant difference among two groups of dengue (OR 1.19, 95% CI 0.92–1.54, P = 0.18) (Rubio et al. 2008; Figueiredo et al. 2010; Pang et al. 2012; Mahmood et al. 2013; Teixeira et al. 2015). The sensitivity analysis illustrated that the removal of Pang et al. (2012) resulted in significantly higher rate of asthma in DHF compared to DF patients (OR 1.47 95% CI 1.06–2.04, P = 0.02). Pruritus was investigated in two studies (Cordeiro et al. 2007; Rubio et al. 2008) comparing DHF with DF and revealed no significance (OR 2.06, 95% CI 0.87–4.89, P = 0.1). Similarly, the comparison of non-specified allergy across DHF and DF illustrated no significant difference in the two groups (OR 0.49, 95% CI 0.05–4.57, P = 0.53) (Figueiredo et al. 2010; Teixeira et al. 2015). One study representing two data sets (Teixeira et al. 2015) provided a comparison between DHF and DF in both food and skin allergy, and the pooled odds ratios revealed no significance in both outcomes (P = 0.05). Three studies (Vazquez et al. 2005; 2014; Bachal et al. 2015) compared IgE positivity between DHF and DF, the fixed effect model indicated significant association between DHF and IgE positivity (OR 4.44, 95% CI 1.43–13.77, P = 0.01). The pooled standardized mean difference from studies comparing dengue-specific IgE level between secondary DF and DHF indicated no significant difference between the two groups in IgE level (P = 0.95) (Fig. 4A) (Vazquez et al. 2005, 2014; Bachal et al. 2015).
-
The pooled standardized mean difference from studies comparing dengue specific IgE level between primary and secondary DF showed a significantly higher level of IgE in secondary DF (P < 0.01) (Fig. 4B) (Vazquez et al. 2005; Vazquez et al. 2014; Bachal et al. 2015).
-
Non-meta-analyzed data were illustrated in (Table 2) which consisted of two studies. Regarding the denguespecific IgE, Koraka et al. (2003) reported significant difference between primary and secondary dengue infection, IgE reference serum and either DHF or DSS, in addition to normal healthy control and either DHF or DSS. However, Míguez-Burbano et al. (1999) showed that there were differences in IgE serum concentration in corresponding to the changes in the status of serum IgM and IgG.
Study Allergic syndrome Compared groups Name N Value Name N Value Pvalue Koraka et al. (2003) in Indonesia (Values of total IgE and dengue-specific IgE in each groups were not described) Dengue-specific IgE Primary Dengue Infection 41 NR Secondary Dengue Infection 127 NR < 10-9 "IgE reference serum" 8 NR DHF 30 NR 0.03 "IgE reference serum" 8 NR DSS 67 NR 0.002 "Non-dengue febrile illness and healthy controls" 34 NR DF 71 NR 0.4 "Non-dengue febrile illness and healthy controls" 30 NR DHF 34 NR < 0.0001 "Non-dengue febrile illness and healthy controls" 34 NR DSS 67 NR < 0.0001 Total IgE (IU/mL) "Non-dengue febrile illness and healthy controls" 34 NR DF 71 NR 0.8 Total IgE (IU/mL) "Non-dengue febrile illness and healthy controls" 34 NR DHF 30 NR 0.539 Total IgE (IU/mL) "Non-dengue febrile illness and healthy controls" 34 NR DSS 67 NR < 0.0001 Míguez-Burbano et al. (1999) in Colombia Value : Mean ± SD Total IgE (IU/mL) Dengue IgM sero (+) 55 315.4 ± 19.9 Dengue IgM sero (-) 113 261.6 ± 5.9 0.016 Dengue IgG sero (+) 136 288.4 ± 5.9 Dengue IgG sero (-) 32 192.6 ± 9.3 0.01 Dengue IgG (+); IgM (-) NR 271.9 ± 8.5 Dengue IgG (-); IgM (-) NR 192.6 ± 34 0.01 Dengue IgG (+); IgM (+) NR 318.7 ± 1.3 Dengue IgG (-); IgM (-) NR 192.6 ± 34 0.01 Dengue IgG (-); IgM (+) NR 329.3 ± 1.3 Dengue IgG (-); IgM (-) NR 192.6 ± 34 0.01 NR not reported. Table 2. Summary of the studies that highlight the association of allergic symptoms with dengue severity that were not eligible for the meta-analysis.