Table 3. Studies Assessing the Effect of Maternal HBV Sero-status on Persistence/Prevalence of HBeAg in a Systematic Review, up to 2012.
| Crude | Adjusted | ||||||||||
| First Author, Year, Region (Reference No.) | Study Design | Age Range, Yearsa | Type of MaternalHBV Sero-marker | Frequency in Exposed Group | Frequency in Non-exposed Group | OR orRRb | 95% CI | P value | OR orHRc | 95% CI | P value |
| 1. Persistence of HBeAg | |||||||||||
| Kojima, 1985, Japan [25] | Co | 2–12 | HBsAg at entry | 7/9 (78%) | 9/28 (32%) | 2.4 | 1.3, 4.6 | 0.02 | N/R | ||
| Kojima, 1985, Japan [26] | Co | 19–48 | HBsAg at entry | 4/4 (100%) | 2/9 (22%) | 4.5 | 1.3, 15.3 | 0.02 | N/R | ||
| Chang, 1989, Taiwan [27] | Co | 0–15 | HBsAgd | 121/142 (85%) | 52/75 (69%) | 1.2 | 1.0, 1.5 | 0.006 | N/R | ||
| Tseng, 2011, Taiwan [28] | Co | 0–16 | HBsAg at entry | 56/137 (41%) | 8/48 (17%) | 2.5 | 1.3, 4.8 | 0.002 | 1.2e | 0.8, 1.8 | 0.5 |
| HBeAg at entry | 41/80 (51%) | 23/105 (22%) | 2.3 | 1.5, 3.6 | <0.0001 | 1.8e | 1.1, 2.8 | 0.01 | |||
| 2. Prevalence of HBeAg | |||||||||||
| Wheeley, 1989, UK [29] | CS | 0–16 | Prenatal HBeAg | 32/42 (76%) | 0/1 (0%) | 9.3f | 0.4, 245.6 | 0.1 | N/R | ||
| Habu, 1991, Japan [30] | CS | Children & adults | HBsAg at entry | 71/101 (70%) | 96/152 (63%) | 1.4 | 0.8, 2.4 | 0.2 | 1.6g | 0.9, 2.9 | 0.1 |
| Tai, 1999, Taiwan [31] | CS | >15 | HBsAg at entry | 67/221 (30%) | 35/131 (27%) | 1.2 | 0.7, 1.9 | 0.5 | N/R | ||
| Hopkirk, 2000, New Zealand [32] h | CS | Children & adults | HBsAgd | 160/281 (57%) | 214/530 (40%) | 2.0 | 1.5, 2.6 | <0.0001 | 1.8g | 1.3, 2.4 | 0.0005 |
| Soderstrom, 2002, Sweden [33] | CS | 2–18 | HBeAgd | 15/16 (94%) | 11/17 (65%) | 8.2 | 0.8, 400.8 | 0.09 | N/R | ||
Abbreviations: ALT, alanine transaminase; CC, case-control study; CI, confidence interval; Co, cohort study; CS, cross-sectional study; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HCC, hepatocellular carcinoma; HR, hazard ratio; N/R, not reported; OR, odds ratio; RR, risk ratio.
When age range was not available, study was categorized as children (<20 years old), adults (≥20 years old) or both children and adults.
Odds ratios for case-control or cross-sectional studies and risk ratios for cohort studies are presented.
Except for the study of Tseng et al. [28] which presented hazard ratios, odds ratios are presented.
When the measurement of maternal sero-status was performed is not known.
Multivariable model included maternal HBsAg, maternal HBeAg, peak ALT and HBV genotype.
As a contingency table contains a zero cell, 0.5 was added to each cell.
Adjusted for age.
The author provided the raw data to compute odds ratio.