General paediatric research: Abstract summaries and commentaries

As discussed by Dr Gold (pages 385–387), it is encouraging that so many outstanding researchers are devoting their skills and expertise to addressing issues that will ultimately improve the health of Canada's children and youth – a noble cause indeed. The major challenge in selecting significant publications from 2003 that will likely impact on child and youth health stems from both the large number and the high quality of the Canadian research reported last year. Given this daunting but exhilarating challenge, I have selected only a small minority of the reports that our search identified and have attempted to focus on those that have significant public health implications and/or appear to represent new information that will likely be of importance in the day-to-day care provided to our children and youth. Unfortunately, space and time do not permit the inclusion of many additional important papers published by Canadian researchers. I apologize both to the reader and the authors for not including more papers in this review.

OBESITY

The alarming and well-recognized epidemic of obesity among Canadian children, youth and adults has already had many adverse health effects and a number of studies have addressed this very important public health issue, including two studies that are reviewed below.

Tremblay MS, Willms JD. Is the Canadian childhood obesity epidemic related to physical inactivity? Int J Obes Relat Metab Disord 2003;27:1100–5.

Tremblay and Willms, from Saskatoon and Fredericton, studied the physical activity and body mass index (BMI) of 7216 children aged seven to 11 years who participated in the 1994 National Longitudinal Survey of Children and Youth. The data in the National Longitudinal Survey of Children and Youth included parental reports of height and weight, which were used to calculate BMI. Children with calculated cut-off BMIs (BMI>25 kg/m²) were classified as overweight and those with a BMI greater than 30 kg/m² were classified as obese. Parental responses were used to identify the amount and type of the children's physical activity. Both overweight and obesity were more common among children from families with lower socioeconomic status (SES) and among those from single parent families. When SES and family structure data were controlled, BMI was still significantly higher among the less active children. Both organized and unorganized sport participation was negatively associated with both overweight (10% to 24% reduced risk) and obesity (23% to 43% reduced risk), while television watching and video game playing were positively associated with being overweight (17% to 44% increased risk) and being obese (10% to 61% increased risk). Watching more than 2 h/day of television was significantly associated with an increased risk of both overweight and obesity. Physical activity and sedentary behaviour were partially due to the effects of SES and family structure, but still remained significant when SES and family structure were considered. The authors also documented that participating in art and dance was negatively associated with both overweight and obesity.

Trudeau F, Shephard RJ, Bouchard S, Laurencelle L. BMI in the Trois-Rivières study: Child-adult and child-parent relationships. Am J Human Biol 2003;15:187–91.

Trudeau et al, from Trois-Rivières and Toronto, studied the relationship between parental BMI and adult BMI in a cohort of 114 French Canadian children measured at 10 to 12 years of age from 1970 to 1977 and, again, at follow-up from 1996 to 1998 (mean age at second measurement was 34.9 years). The children had been enrolled in the Trois-Rivières semilongitudinal study of growth and development and were compared with their parents. BMI increased as expected during the study interval, and, as adults, the children had slightly greater BMIs than did their parents at a comparable age. Regression analysis of adult BMIs related statistically only with childhood BMIs but not with parental BMIs. The study was not designed to determine the relative contributions of muscle mass compared to fat in the adult BMIs, but it was clear that childhood BMI was a much better predictor of adult BMI than parental BMI.

Comment: Both the Tremblay and Willms and the Trudeau et al studies address an increasingly important public health concern regarding the predictors and probable causes of adult obesity. The Trudeau et al study suggests that while genetic factors are likely important in the etiology of obesity, parental BMI when the children are 10 to 12 years of age may not be a reliable predictor of adult BMI and, by inference, obesity. However, BMI during childhood was a significant predictor of adult BMI. Whether a similar study done in 2004 would confirm this relationship is unknown, but Tremblay and Willms' data strongly suggest that a modifiable determinant of obesity is the amount of physical activity done by seven- to 11-year-old children. Clearly, increased physical activity and decreased sedentary behaviour, including television watching and video game use, are practical and potentially achievable goals for paediatricians and family physicians to recommend and monitor in their paediatric patients. It is clear that obesity and overweight can, and do, begin in childhood. The increasing burden of obesity among Canadian children and youth supports the need for child health practitioners to advocate physical activity for their own patients as well as in their community leadership roles.

SOCIOECONOMIC FACTORS AND CHILD HEALTH

Langille DB, Curtis L, Hughes J, Murphy GT. Association of socio-economic factors with health risk behaviours among high school students in rural Nova Scotia. Can J Public Health 2003;94:442–7.

Langille et al from Halifax surveyed grade 10 to 12 students in four northern Nova Scotia high schools to determine the association between SES and a number of variables, including substance abuse, sexual behaviour and suicide attempts. The surveys were done by trained teachers during class time in 2000. Seventy-nine per cent of eligible students participated in the study (2198 students, 48% males, 52% males, age range of 14 to 20 years). The data revealed that almost 25% of students smoked, 19% of males and 8% of females had used marijuana more than 10 times in the past month, over 30% had more than five drinks on at least one occasion in the previous month, more than 10% had intercourse before age 15, and 3% of males and 4% to 8% of females had attempted suicide. Drinking and driving was reported by over 14% of grade 12 males. Among both males and females, smoking was the behaviour most associated with a lower SES. Both parental education and employment were found to be related to substance abuse, with those mothers with postsecondary education having fewer children smoking, while the father's education was similarly related to decreased heavy drinking behaviour. Adolescents living with single parents were more likely to smoke, use marijuana and have early intercourse. The only SES variable associated with suicide attempts was lower maternal education.

Séguin L, Xu Q, Potvin L, Zunzunegui MV, Frohlich KL. Effects of low income on infant health. CMAJ 2003;168:1533–8.

In 1998, Séguin et al from Montreal interviewed 2223 mothers of five-month-old (range 15 to 36 weeks) Quebec infants participating in the Quebec Longitudinal Study of Child Development. Premature infants were not included. As part of this population-based study, the authors reviewed the infants' health records to determine their neonatal health status and asked the mothers about their baby's health. Statistics Canada's low-income thresholds were used to determine income level. A family that devotes 20% more of its before-tax income to food, shelter and clothing than does the average family was considered low-income. The income used to indicate a “moderately inadequate” income for a family of three was between $16,238 and $27,063, and a “very inadequate” income was less than $16,238. The participation rate of eligible families was 83.1%, although the final rate fell to 75.8% when nonresponders for some of the data sets were excluded. The lowest response rates were from mothers with the lowest levels of education and those who spoke neither French nor English. Twelve per cent of infants lived in households with moderately inadequate family incomes and 15.5% lived in households with very inadequate family incomes. Over 31% of mothers with moderately or very inadequate incomes perceived their baby's health to be less than excellent compared with 20% of mothers with sufficient income. Significantly more mothers with low incomes reported chronic health problems than did mothers with sufficient income. Infants of mothers with moderately inadequate income were admitted to hospital since birth significantly more often (21%) than were infants of mothers with sufficient income (11%) or very inadequate income (13%). Using stepwise regression multivariate analysis and adjusting for infant characteristics and neonatal health, mothers with inadequate incomes still perceived that their infants had less than excellent health, and those with moderately inadequate incomes, but not very inadequate incomes, had significantly more admissions to hospital. The factors leading to the significantly increased admissions to hospital in the infants of those families with moderately inadequate incomes but not those from very inadequate incomes were not clear.

Comment: The Langille et al and Séguin et al studies demonstrate that social factors and low income impact on child health, during both infancy and adolescence. While these results are neither surprising nor unexpected, they highlight the negative effects that social factors such as low parental education and, especially, family poverty have on child and youth health. The high rate of drinking and driving among Nova Scotia teenagers was also demonstrated in Ontario by Adlaf et al (1). The Langille et al and Adlaf et al studies highlight the adolescent component of drinking and driving, which like obesity, is a major public health problem, and physicians who care for children and youth may play an important role in its prevention.

Other Canadian publications last year identified the negative impact of poverty on the availability of medications for children with significant disease (2) and the effect of maternal education, and to a lesser extent, low income, on infant breastfeeding (3). The effect of low family income on hospital admissions reported in the Séguin et al study remains unexplained but is consistent with the report from Alberta by Sin et al (4) of the impact of family income on emergency department use by children with asthma from low income families.

When taken together, these papers and others demonstrate the huge effects of social factors on health, particularly child and youth health. Canadian researchers have provided clear evidence of some of the negative impacts on child health of poor parental education and poverty. The challenge for all of us involved in child and youth health is how best to intervene to prevent or at least ameliorate these negative effects.

FOLATE INTAKE: PREVENTION OF NEURAL TUBE DEFECTS AND POSSIBLY MORE

French MR, Barr SI, Levy-Milne R. Folate intakes and awareness of folate to prevent neural tube defects: A survey of women living in Vancouver, Canada. J Am Diet Assoc 2003;103:181–5.

French et al from Toronto and Vancouver interviewed a randomly selected cohort of Vancouver women between the ages of 18 and 45 years to determine their folate intake, and their knowledge and awareness of the relationship between folate intake and neural tube defects (NTD). Only women fluent in English were eligible, and those who were pregnant or had been pregnant in the previous six months were excluded. One hundred forty-eight women completed the study, which represented 35.7% of the households contacted by the investigators. Eligible women who refused to complete the survey were significantly more likely to have children and less likely to have heard of folate than those who completed the survey tool. Ninety-nine per cent of the women had completed high school, 69% had annual household incomes greater than $40,000, 65% were married/common-law and 56% had children. Folate intake was directly related to the presence of folate fortification of bread and grain products. In the absence of fortification, 64% of women fell below the estimated average daily folate requirement, but with bread and grain fortification, only 23% fell below the requirement. Even with folate supplementation, 14% of the women still fell below the daily requirement. In contrast, 95% of women had heard of folate but only 36% knew that it was important during pregnancy. Only 20 women (30%) thought folate supplements should be taken before pregnancy to reduce the risk of birth defects. Women who had children were significantly more aware of the need for folate during pregnancy than were women who had no children. Most women (78%) indicated that they would be willing to take folate supplements during pregnancy if it was demonstrated that it would reduce the risk of NTD. A large majority of women (94%) indicated that their family physician would be a desirable source of information about folate. Almost 90% of the women who had discussed folate with their physician were aware of the relationship between folate and health, compared with only 68% of those who had not discussed the issue with their physician.

Comment: Given the well-documented benefit of preconception folate intake in the prevention of NTD and other congenital anomalies, the French et al study is both encouraging and worrisome. Indeed, the incidence of NTD continues to decline in Canada, an effect that is likely related, at least in part, to the widespread use of prenatal folate supplementation. It is encouraging that those women who had discussed the issue with their family doctors were knowledgeable and that most women would likely take a supplement if they knew the evidence of its benefit. This is another obvious opportunity for physicians to intervene effectively on a public health issue that has significant benefit to society in general and child health in particular. However, the observation that even with bread and grain folate fortification, a significant number (23%) of women do not consume the required amount of folate that would be expected to decrease the risk of NTD is somewhat discouraging. Another study by French (5) from Ontario suggests that a previously unrecognized additional benefit of folate fortification may be a decrease in neuroblastoma, the second most common solid tumour occurring during childhood. Neither the rates of acute lymphoblastic leukemia nor hepatoblas-toma in childhood appeared to be related to folate fortification. While the possible effect of folate on the rates of childhood cancer remains to be verified, it is well established that folate has remarkable benefits in preventing NTD, and therefore, all physicians involved in the care of women who are in their reproductive years must continue to extol the benefits of adequate folate intake, ideally, before conception.

RISK OF ACUTE LYMPHOBLASTIC LEUKEMIA: GENES AND RADIATION

Infante-Rivard C. Diagnostic x rays, DNA repair genes and childhood acute lymphoblastic leukemia. Health Phys 2003;85:60–4.

This case-control study attempted to determine the importance of x-rays in the development of acute lymphoblastic leukemia (ALL) in Quebec children. Parents of 701 children under 14 years of age diagnosed with ALL between 1980 and 1998, and matched controls (age, sex, region of residence) were asked about the number of nondental x-rays their children had received. A single x-ray did not increase the risk of ALL (odds ratio [OR] 1.13, 95% CI 0.84 to 1.50), whereas two or more x-rays increased the risk (OR 1.47, 95% CI 1.11 to 1.95). The types of radiographs were similar in both cases and controls. In the second part of this study, 207 children with ALL were genotyped for polymorphisms in DNA repair genes, which may be important in repairing the genetic damage done by x-rays. Six allelic variants of four DNA repair genes were determined and their possible association with x-ray exposure was assessed. Most of the polymorphisms appeared not to be associated with ALL; however, a polymorphism in the APE gene base excision repair system was suggestive of an increased risk among boys and a reduced risk among girls. There was an indication that a mutation in the hMLH1 mismatch repair gene may lower the risk of ALL in girls.

Comment: ALL is the most common cancer in children and its etiology is almost surely a combination of genetic and environmental factors. Infante-Rivard's study suggests that children who have more than one nondental x-ray may have a slightly greater risk of developing ALL than children who have only one or no x-rays. The OR was only modestly increased and, although statistically significant, the relative importance of x-rays in the development of ALL remains uncertain. The association of x-ray exposure and DNA repair gene polymorphisms can be viewed as additional indirect evidence that radiation exposure may indeed be a risk factor for ALL, and perhaps individuals with certain genotypes may be at higher risk than others. It would be useful if future research could identify children whose genotype puts them at increased risk of ALL following x-rays or other environmental exposures, potentially providing opportunities to lower this risk in these children.

In this regard, other Quebec authors have addressed the possible role of DNA repair genes and the risk of childhood ALL (6). The study by Mathonnet et al (6) suggested that the hMLH1 mismatch repair gene might indeed play a role in the etiology of ALL. These data are preliminary but highlight the progress being made in the quest to unravel both the environmental and genetic factors that are important and interactive in the development of childhood ALL. While we are not yet at the stage where the understanding of the complex environmental genetic interactions can lead to specific interventions to decrease the occurrence of ALL, these studies are consistent with the recommendation to minimize the amount of x-ray exposure children receive.

IMMUNIZATION

Patrick DM, Bigham M, Ng H, White R, Tweed A, Skowronski DM. Elimination of acute hepatitis B among adolescents after one decade of an immunization program targeting Grade 6 students. Pediatr Infect Dis J 2003;22:874–7.

This study from Vancouver assessed the effectiveness of British Columbia's hepatitis B (HB) universal immunization program introduced in 1992. At the time, British Columbia had the highest rate of HB infection in Canada at 33.7 cases per 100,000 compared with the national average of 9.9 cases per 100,000. The program initially involved immunization of grade 6 students, and in 2001, universal infant HB immunization was added to the program. The authors documented annual vaccine uptake and HB disease rates using data from the British Columbia Centre for Disease Control. Results indicated that immunization coverage ranged from 90% to 93% of grade 6 students for each year between 1993 and 2001. Both the number of cases of HB infection and the rate per 100,000 population declined significantly from 1992 to 2001, when the study concluded. The HB rate among the cohort of adolescents and young adults eligible for vaccination fell from 1.7 to 0 per 100,000. There were no cases of HB in this age group in 2002. The program had differing effects in urban and rural areas, with urban areas having only a 6.4-fold decline in acute HB infection (95% CI 2.8 to 14.7) compared with a 49-fold reduction in rural areas (95% CI 5.4 to 435).

Comment: This study clearly documents the declining rate of HB infection in British Columbia since the introduction of the universal immunization program. In addition, the authors document the virtual elimination of acute HB infection in adolescents and an approximately 50% reduction in the rate in the whole population. British Columbia was the first province to introduce universal HB vaccination, and subsequently, all other Canadian provinces have introduced vaccination programs. The data in this report support the development of programs in other jurisdictions. The authors note that other programs in British Columbia involving immunization of high-risk populations (eg, injection drug abusers) and other strategies must also be important factors in the decline of HB in the general population. This study also suggests that the program was more effective in rural areas than in urban areas where the burden and number of HB infections is greatest – indicating that additional measures are needed to decrease the impact of HB on society. The striking decline in HB infection in children and adolescents is particularly noteworthy and is another example of the benefit of childhood immunization programs implemented at the population level.

de Courval FP, De Serres G, Duval B. Varicella vaccine: Factors influencing uptake. Can J Public Health 2003;94:268–71.

de Courval et al from Quebec City addressed the low use of varicella vaccine among Quebec children despite the observation that a majority of parents informed of the complications of varicella indicated that they were interested in the vaccine for their children. Between July and October 2000, phone interviews were used to collect data about a number of demographic variables as well as the varicella vaccination status of 14- to 17-month-old children in the Quebec City immunization registry, which includes over 98% of the children in the region. The parents of 74% (663 of 893) of the eligible children were successfully contacted and over 87% of these agreed to participate in the study. Although 66% of parents were aware of the varicella vaccine, only 37% reported that their usual vaccine provider offered the vaccine (45% if it was a paediatrician but only 29% if the provider was a public health nurse or a general practitioner). Those with higher-income parents and higher maternal education level were significantly more likely to have been offered the vaccine. Among those offered the vaccine, 33% were vaccinated, with significantly more children receiving the vaccine among those with older mothers and from higher-income families. In contrast, the type of vaccine provider (paediatrician, public health nurse or general practitioner) did not differ among those who chose to have their children vaccinated. The more information the vaccine provider gave the parents, the more likely the children received the vaccine. Information on vaccine safety was significant. Thirty-nine per cent of parents who refused the offer of vaccination cited the cost of the vaccine as the reason for their refusal, while 35% indicated that it was the vaccine provider's weak recommendation. Among those reportedly not offered the vaccination, 58% indicated that they would have been interested in the vaccine, but 43% became “uninterested” when told of the price of the vaccine ($80). More of those with higher incomes remained interested in vaccination after being informed of the cost. Among those parents who were not interested in vaccination when told of the vaccine cost, 61% became interested when informed of the complications of varicella infection.

Comment: In 1999, the National Advisory Committee on Immunization recommended varicella vaccination for children older than 12 months, as well as others at risk of varicella infection, but in most jurisdictions, the cost of varicella vaccination is not included in provincial vaccination programs and the cost must be borne by the patient or his/her family. The de Courval et al study clearly indicates that the cost of varicella vaccination is a deterrent to vaccination, especially for lower-income families. Low family income impacts negatively on many aspects of child health and, not surprisingly, makes children from low-income families less likely to receive the benefits of varicella vaccination. In addition, it is somewhat surprising that reportedly, only 37% of the children's usual vaccine provider offered varicella vaccination. Offering the vaccination appears to be more effective when the provider also provides more information, particularly information on the complications of varicella infection. Paediatricians appeared to be modestly more likely to offer vaccination than were family physicians, but in both physician groups, less than 50% reportedly offered the vaccine. Hopefully, in 2004, many more vaccine providers, whether paediatrician, family physician or public health nurse, are offering varicella vaccination to the children they care for. Varicella vaccine cost and its absence from some provincial vaccination programs presents yet another opportunity for physicians to advocate provincial public health programs aimed at protecting our children. Although the National Advisory Committee on Immunization recommends universal varicella vaccination, there are clearly a number of obstacles that must be overcome before the recommendation can be realized.

Grant VJ, Le Saux N, Plint AC, et al. Factors influencing childhood influenza immunization. CMAJ 2003;168:39–41.

Grant et al examined the factors that affected influenza vaccination among a convenience sample of 203 children older than six months. The children were recruited from those presenting to the Children's Hospital of Eastern Ontario's emergency department between January and March 2001, when the province of Ontario sponsored publicly funded influenza immunization. Parents of eligible children were given a standard questionnaire, which addressed factors influencing the decision to have their children immunized. Only 27% of the children had been immunized against influenza. Children who had been immunized did not differ in age, sex, daycare attendance, parental education level or emergency department diagnosis from those who had not been immunized. Children who had family members immunized against influenza (relative risk [RR] 7.5, 95% CI 3.7 to 15.1) or who had been previously immunized (RR 2.7, 95% CI 1.8 to 2.4) were more likely to have had influenza vaccination. The majority of parents incorrectly thought that vomiting and diarrhea were symptoms of influenza. Parents of nonimmunized children were significantly more likely to believe that influenza-like illness was a side effect of vaccination, that the side effects were worse than infection and that immunization would weaken the immune system. The most common reasons reported for immunization were prevention of influenza and physician recommendation. Fifty-seven per cent of parents reported discussing immunization with their child's physician, but only a small majority reported that vaccination was recommended. When immunization was not recommended, 96% of children were not immunized. Logistic regression analysis revealed that two factors were significantly associated with vaccination, having a child with chronic illness and discussing influenza immunization with the child's physician.

Comment: This study is consistent in many ways with the report of de Courval et al which identified some of the factors that influence parents to have their children vaccinated. While universal varicella vaccination has been recommended, universal influenza vaccination has not yet received such a recommendation, even though it was publicly funded and recommended for all children over six months of age at the time of Grant et al's study in Ontario. The National Advisory Committee on Immunization's 2003–2004 statement on influenza vaccination states that “healthy adults and children should be encouraged to receive the vaccine” (7). There is an increasing number of reports that suggest there is a benefit for influenza vaccination in children who are not high risk, but at present, individual physicians and families will have to determine the benefits and risks of immunizing non-high-risk children older than six months without the benefit of definite national recommendations. While we await such recommendations from national agencies, both Grant et al and de Courval et al's studies demonstrate the importance of discussion of immunization by physicians with parents and the need to provide information about the benefits of the vaccine. In addition, the cost of nonpublicly funded vaccination is clearly an impediment to vaccination. Influenza vaccination, like varicella vaccination, is not yet universally covered by provincial plans, but physicians can and should influence the rate of childhood immunization, and therefore, the burden of these infectious diseases on their young patients and families. Physicians have the opportunity and the obligation, at least in my view, to provide the current information on the risks and benefits of vaccines and to offer vaccination to their patients.

HUMAN METAPNEUMOVIRUS: HERE IN CANADA

Bastien N, Ward D, Van Caeseele P, et al. Human metapneumovirus infection in the Canadian population. J Clin Microbiol 2003;41:4642–6.

Bastien from Manitoba and her colleagues from three other Canadian centres conducted a study on the prevalence of human metapneumovirus (MPV) in four provincial public health laboratories in an effort to determine the impact of this recently recognized respiratory virus on acute respiratory tract infections in Canada. They reviewed MPV isolates obtained during the 2001/2002 (October to April) influenza season and the clinical characteristics of the infected patients. MPV was detected by molecular methods (reverse transcriptase polymerase chain reaction) and 445 specimens were tested. MPV was detected in 14.8% of all specimens and identified in all participating provincial laboratories. Patients of all ages were infected. Twenty-eight per cent of those zero to five years, 4% of those six to 10 years, 10% of those 11 to 20 years, 25% of those 21 to 50 years and 31% of those older than 50 years were infected, with males and females having similar rates. MPV infections peaked in February (41% of isolates) and March (39%). Clinical symptoms included fever (57%), cough (54%), influenza-like illness (45%) (headache, nausea and muscle ache occurred in less than 10% of patients). Diagnoses included rhinitis (27%), pneumonia (18%), bronchiolitis (13%) and bronchitis (4%). Thirty-three per cent of patients were hospitalized, with significantly higher rates of admission found among patients younger than five years (58%, P=0.0005) and older than 50 years (45%, P=0.0044). Those admitted to hospital also had the highest rates of pneumonia, bronchiolitis and bronchitis. The data demonstrate that MPV infection was widespread in Canada during the winter of 2001/2002. This was a time when other respiratory viral infections were also common, with many having similar symptoms and associated diagnoses.

Comment: MPV was discovered in 2001 in the Netherlands and Bastien et al have now demonstrated that this newly recognized respiratory virus appears to be a significant cause of acute respiratory infections among Canadians of all ages, with particularly common and severe infections in those younger than five years and older than 50 years. While more research is necessary to fully define the burden of MPV in Canada, it is likely that this virus is a significant cause of many viral lower and upper respiratory infections during the winter months, particularly those that are not attributable to influenza and respiratory syncytial virus, a very common cause of viral bronchiolitis in infants. Reports from around the world clearly documenting the worldwide distribution of MPV and its predilection for the young and elderly have now emerged. More recent reports from the United States and elsewhere suggest that MPV infection has a very similar impact in children, both clinically and epidemiologically, to respiratory syncytial virus and influenza (8–10). Thus, Bastien et al have been at the forefront of defining the impact of a new respiratory pathogen that infects children and adults, and is a significant cause of disease in Canada. Much research still remains to be done to address issues of treatment and prevention of MPV infection. As well, diagnostic tools need to be developed to enable physicians to accurately distinguish MPV from other common viral respiratory tract infections. In the long term, immunization will likely be another important area for further research in MPV infection.

TYPE 2 DIABETES: GENETIC CAUSES

Hegele RA, Zinman B, Hanley AJ, Harris SB, Barrett PH, Cao H. Genes, environment and Oji-Cree type 2 diabetes. Clin Biochem 2003;36:163–70.

In this review paper, Hegele et al from London, Ontario, review and summarize their groundbreaking work on the genetic causes of type 2 diabetes. At approximately 40%, the prevalence of type 2 diabetes in Canadian Oji-Cree is among the highest in the world. Their work has uncovered genetic determinants of Oji-Cree type 2 diabetes and related metabolic traits. The most important genetic discovery was the private G319S mutation in transcription factor HNF-1α, encoded by the HNF1A gene. HNF1A G319S was discovered by candidate gene sequencing and would have been missed using the currently favoured strategy of genome-wide scanning. G319S was associated with increased odds of having type 2 diabetes across the whole study sample and in all subgroups, including adolescent Oji-Cree. Furthermore, G319S had specificity and positive predictive value of 97% and 95%, respectively, for developing type 2 diabetes by age 50. The protein bearing the G319S mutation had impaired function in vitro. Sigmoidal modelling showed that each dose of the G319S allele accelerated the median age of diabetes onset by about seven years. This approach also showed that environment more strongly accelerated the median age-of-onset of Oji-Cree diabetes than did G319S, which could have implications for intervention strategies to reduce the burden of this epidemic. Hegele's work has also provided evidence for genetic determination of related metabolic traits including plasma liproproteins and fatty acid binding protein in the Oji-Cree population.

Comment: As noted above, type 2 diabetes is becoming increasingly prevalent in Canada and is occurring at an alarming rate in children and youth. Lifestyle factors, such as inactivity and increased intake of dietary fat, are clearly major determinants, but as the pioneering work of Hegele's group has demonstrated, genotype plays a crucial role in the etiology of type 2 diabetes among the Oji-Cree and, likely, among other populations as well. The authors review the methodology used to identify the mutations in the transcription factor HNF-1α gene, which dramatically increase the risk of developing type 2 diabetes in those people who have the G319S mutation. The mutation appears to decrease production of the HNF-1α protein, which in turn leads to decreased, but not absent, expression of HNF-1α-dependant genes, including insulin. It appears that in the presence of the mutation, the stress of obesity and hyper-lipidemia greatly increases the risk of type 2 diabetes as well as its early onset. Indeed, when the age of onset of type 2 diabetes was compared between Oji-Cree born before 1935 and those born after 1935, it was found that the more recently born cohort had a much younger age of onset. This observation strongly suggests that the increasing prevalence of obesity and a more sedentary lifestyle are interacting with the mutation, which is unlikely to have changed in frequency since 1935, resulting in the strikingly high rate of type 2 diabetes. The obvious interaction of diet and activity with genotype provides yet another example of the potential for intervention to prevent or delay the onset of type 2 diabetes. Although it is obvious that everyone should be counselled about the importance of healthy eating and exercise, in the future, high-risk genotypes may be identified in specific populations, which could facilitate more selective intervention and counselling.

REFERENCES