Review Article

Paternal Smoking and Risk of Childhood Acute Lymphoblastic Leukemia: Systematic Review and Meta-Analysis

Table 1

Description of the 18 original research studies on paternal smoking and childhood ALL included in the meta-analysis.

StudyCases/controlsAge (years)Case recruitmentControl selectionOverall*Before pregnancyDuring pregnancyAfter birth

Lee et al. 2009, Korea [42]106/1640–18Incident childhood leukemia cases diagnosed in three hospitals in Seoul between 2003 and 2005Other patients from the three hospitals where cases came from, matched for age and sex≥400 cigarettes/life timenumber of PYs, ≤10 or >10 PYssmoking at home during pregnancynumber of PYs, ≤10 or >10 PYs
Rudant et al. 2008, France [33]647/1681<15Cases were identified directly by investigators with support of the national cancer registry in France between 2003 and 2004Population based, with quota match for age and sexCPDs from the year prior to the child's birth to the interviewCPDs from the year prior to the child's birth to the interviewCPDs from the year prior to the child's birth to the interview
MacArthur et al. 2008, Canada [28]351/3990–14 Incidence case from 5 regions in Canada, diagnosed between 1990 and 1994From health insurance roll, matched for age, sex, and area for each caseever smokerCPDs before pregnancyCPDs in the year prior to the child birth
Menegaux et al. 2007, France [30]407/567<15Cases derived from the national registry in 14 regions between 1995 and 1998Population based, frequency match for age, sex, and areaCPDs in the 3 months before pregnancyCPD from the child's birth to the diagnosis
Chang et al. 2006, USA [23]228/306≤15hospital diagnosed cases between 1995 and 2002, North California Childhood Leukemia StudyRandom selection from birth certificates, individual match for age, sex, and maternal raceever smoker: ≥100 cigs before diagnosisCPD in the 3 months before pregnancy
Menegaux et al. 2005, France [31]240/142<15newly diagnosed acute leukemia cases from 1995 to 1999 in four cities in FranceMostly from departments of orthopedic of the same hospital, matched for age rangeCPDs from the index birth to interview
Pang et al. 2003, England [32]1375/6987<15National wide population-based cancer cases diagnosed by regional oncology units between 1991–1994 in Scotland and 1992–1994 in EnglandRandomly selected from Family Health Serves Authorities lists, and matched for sex, date of birth, and geographical area of residenceever smoked before conception
Sorahan et al. 2001, England [37]139/132<15Children first diagnosed with leukemia in 3 areas in England in 1980–1983§From General Practitioners list, matched for sex and date of birthCPDs
Infante-Rivard et al. 2000, Canada [26]486/4860–9Cases from tertiary care centers for childhood cancers, diagnosed in 1980–1993, QuebecPopulation based from family allowance, matched for age, sex, and areaCPDs between birth and date of diagnosis
Brondum et al. 1999, USA [21]1618/1986<15Newly diagnosed with leukemia via clinical trial registries from 1989 to 1993, CCG studyRDD, individually matched on age, race, area code and exchangesmoking amounts during lifetimeever smoked one month before pregnancyever smoked during the three trimestersever smoked during nursing period
Schuz et al. 1999, German [43]686/2588<15From a national wide cancer registry (1992–1997) and from cases diagnosed (1980–1994) and lived in vicinity of nuclear installationsrandomly selected from complete files of local offices of registration of residents, matched for area, sex, and similar date of birth (within one year)CPDs in the last 3 months before pregnancy
Sorahan et al. 1997b, England [38]573/573<16Children who died from leukemia in England, Wales, and Scotland between 1971 to 1976From birth registers of local authority areas where cases died, matched by sex and date of birthcurrent status, 6 levels from 0 to 40 CPD
Sorahan et al. 1997a, England [36]367/367<16Children who died from leukemia in England, Wales, and Scotland between 1953 to 1955From birth registers of local authority areas where cases died, matched for sex and date of birthcurrent status, 4 levels from 0 to 20 CPD
Ji et al. 1997, China [41]114/114<15Newly diagnosed childhood cancer cases from 1985 to 1991 in ShanghaiPopulation-based controls from household registry, matched for sex, and year of birthPYs before conceptionPYs after birth
Shu et al. 1996, USA, Canada [34]191/363≤18 monthsinfants newly diagnosed matched for leukemia from 1983 to 1988 via clinical trial registriesRDD, individually matched for year of birth, telephone area code, and exchange numberCPDs in the month prior to pregnancyCPDs during pregnancy
Sorahan et al. 1995, England [35]371/371<16Children who died from leukemia in England, Wales, and Scotland between 1977 to 1981§From the birth register of the local authority area in which the case child died, matched for sex and date of birthCPDs during prenatal period, categorized into 6 levels
John et al. 1991, USA [27]47/1840–14Incident cases aged 0–14 diagnosed in Denver, Colorado from 1976 to 1983RDD, matched on age, sex, and geographic area.CPDs during the 12 months prior to birthCPDs during the 12 months prior to birth
Magnani et al. 1990, Italy [29]142/3076.1/6.6#Pediatric hospital prevalent cases in Turin Italy, diagnosed between 1974 and 1984Randomly sampled from medical or surgical wards of the same hospitals, no matchesCPDs up to child’s birthCPDs up to child’s birth

RDD: random digit dialing; CPD: cigarettes per day; PY: package year; *: overall status means without specific exposure time period specified; §: There was a small degree of overlap between cases included by Sorahan et al. 2001 [37] and cases included by Sorahan et al. 1995 [35]; #: mean age of cases at diagnosis: 6.1 years, with standard deviation of 3.6 years and mean age of controls: 6.6 years, with standard deviation of 3.5 years.