Discussion
The prevalence of IR observed in our study (42.3%) is very close to that reported by Arshi et al (42.9%) in an Australian population, using the same cut off of HOMA-IR ≥1.77 9. Literature on IR in Indian children is scarce. A recent study reporting on Indian children with asthma, aged 6-18 years, estimated the prevalence to be 15.5% 24. The study had used a HOMA-IR value of 2.5, which has been suggested to be more accurate for the Indian population23. Applying the same threshold, the prevalence of IR in our population comes to 20.5%, which is comparable. Unfortunately, our study did not include a control group. Overall, the median HOMA-IR in our cohort of children with asthma was lower than the average reported in a study conducted in schoolchildren aged 6 to 16 years, in the eastern part of the country 25. However, in the latter study, 28.2% participants were overweight and obese, compared to only 10.8% in our population. In addition, there might be other factors accounting for the difference, like a differential distribution of post-pubertal children, as HOMA-IR values have been shown to increase with increase in sexual maturation 23.
IR has been suggested to be an early biochemical marker of MS, developing at age 18-19 years 26. We, therefore, believe our result to be significant because relying on anthropometry alone, the risk of developing MS in our population would appear to be lower. If the prevalence of IR in Indian children with asthma is confirmed to be as high in larger studies, this might indicate the need to adopt stricter lifestyle measures in early adolescence.
The prevalence of dyslipidemia in our study was 86.7% (79.4-94.0%). The high prevalence was predominantly because of commonly observed low HDL-C in our study subjects, seen in 80.7% (72.2-89.2%). In a population of 695 Indian urban adolescents, in the age group of 10-18 years, low HDL-C was reported in 27.3% 27. One reason for our group to have a low HDL-C could be the higher proportion of males, as it has been shown that boys typically show a sudden decline the HDL-C levels around the pubertal age, compared to females28. We believe it is extremely unlikely that asthma alone would be responsible for such a drastic abnormality.
Only one child in our study fulfilled the criteria for MS. Similarly, in another Indian cohort of 90 children, only two had MS24. Ross et al found 12 children to qualify for presence of MS in a group of 116 children with asthma12. However, contrary to Ross’s study where 52 children (44.8%) were obese, none of the children sampled in our study were obese. This might possibly explain the observed low prevalence of MS in our study. However, as previously stated, we believe that a higher proportion of children, who have IR currently might be at risk for developing MS later.
Studies have shown that among children & adolescents with asthma, IR and MS are associated with higher bronchial hyperresponsiveness and severe asthma 7,29. Also, dyslipidemia has been shown to be associated with higher airway resistance measured by forced oscillation technique 30. We hypothesized that it might be possible to demonstrate an association between metabolic parameters and asthma symptom control. To test this, we compared the groups with controlled, partly controlled, and uncontrolled asthma. While there was no statistically significant difference in median HOMA-IR of the groups, there was a significant increasing trend in serum insulin level, associated with poorer asthma control. There was a similar association seen with presence of IR, defined as HOMA-IR ≥2.5.
Further, we performed a similar comparison for lipid levels and there was a significant increase in serum TG and LDL-C with poorer control. The association of poor asthma control with fasting serum insulin, presence of insulin resistance (HOMA-IR ≥2.5) and serum TG persisted after adjusting for BMI. Simply put, children with asthma who had higher fasting serum insulin, higher serum TG or HOMA-IR ≥2.5 were at a higher risk of having a poorer asthma control, irrespective of their BMI.
These results show that metabolic derangements are directly associated with clinically relevant outcomes like asthma symptom control. To the best of our knowledge, this is the first study demonstrating a correlation between lipid profile and asthma control in children. Unfortunately, the current study design doesn’t permit further delineation between association and causality. It might be worthwhile to explore these metabolic parameters in children with poor asthma control and see if treatment addressing these has a beneficial effect on asthma symptoms.
We also explored correlations between metabolic parameters and spirometry indices (table 3) and found a weak negative correlation between lipid levels and PEFR. The correlation was not significant statistically on adjusting for height. Similar negative correlations with metabolic markers have been reported by a few pediatric studies but the results are not consistent and have often lacked confidence24,31,32. The most likely explanation for the same is that there are several mechanical and non-mechanical factors influencing lung function, and consequently the effect of a single metabolic parameter might not be very strong.
Our study is unique because we have focused on the relationship between metabolic derangements and clinically useful outcomes like symptom control, which are helpful in clinical decision making. Our group included children with physician diagnosed asthma rather than relying on self-reported symptoms. We used population specific standards for anthropometry, blood pressure centiles and spirometry indices.
We acknowledge a few limitations in our study. Of the initially enrolled 97 children, only 83 reported for sampling. There was disproportionate representation of the genders with boys predominating. As there was no control population, prevalence of metabolic derangements could not be compared to that in healthy children. Also, the cross-sectional design prevents us from exploring the temporal association and hence, causality between poor symptom control and metabolic abnormalities.