While pulmonary function testing (PFT) is well established for older children (>5 years) and adolescents, it is still evolving for use in infants and preschool children [1]. Sophisticated equipments are now commercially available, which can measure pulmonary function even in premature babies [2].

This prospective birth cohort study was conducted in the Department of Pediatrics of a tertiary-care institute in Delhi, India from August 2012 to March 2017. The study was approved by the Institutional Ethics Committee. Written informed consent was taken from the parents/guardians. All neonates born at the institute during the study period were screened for eligibility. The inclusion criteria were age £4 weeks, full-term (³37 weeks of gestation), and appropriate for gestational age (weight 10th- 90th centile) babies [10]. The exclusion criteria were any perinatal insults (e.g. birth asphyxia, meconium aspiration, any amount of respiratory distress requiring respiratory support, pathological hyperbilirubinemia or seizure), known major congenital birth defect, required parenteral antibiotic or fluid, neonatal cholestasis, chronic kidney disease or inborn error of metabolism, and mother having antepartum or postpartum hemorrhage, preeclampsia or eclampsia, HIV infection, or parents’ refusal for regular follow-up for three years. The enrolled babies were followed-up every 6 months (±8 weeks) and also whenever they had an acute respiratory infection or any other acute condition. The babies were clinically examined, and anthropometric measurements (weight, length, head circumference) were recorded at enrolment and each visit.

At enrolment and 6 months of age, IPFT were performed either in awake or natural sleep state after breastfeeding. Beyond six months of age, syrup Triclofos (25-50 mg/kg/dose) was used for sedation, whenever required. The maneuvers were performed in the supine position. Baseline TBFVL was completed within four weeks of birth, while RTC and RVRTC were first performed once the child was ³8 kg as per recommendations of the manufacturer of IPFT equipment. IPFT were repeated at every follow-up visit. The IPFT indices and their physiological interpretation are described in Web Table I.

Statistical analysis: Clinical information during each visit was recorded manually into case record form; data were then entered into Microsoft Access software. IPFT data were automatically stored in software (SPIROWARE, Eco Medics AG, Duernten, Switzerland) after each procedure; data of each individual were extracted and managed in Microsoft Access software. Data were analyzed using Stata software v.13 (Stata Corp, College Station, TX, USA). Quantitative variables were summarized using mean and standard deviation if normally distributed; for skewed distribution, median (interquartile range) was used. Chi-square test was used for the analysis of categorical data. The change in IPFT indices from birth to 36 months of age were calculated with the Wilcoxon sign-rank test. Comparison of IPFT indices between gender and other parameters was analyzed with Wilcoxon rank-sum test (Mann-Whitney Test). A P-value of <0.05 was considered significant. For multivariate analysis, mixed-effects linear regression analysis was performed; IPFT indices were taken as the dependent variable, individual subjects as the random effect and gender, weight, length, and age as fixed effect. The LMS chartmaker Pro (Medical Research Council, UK) was used to model the expected median (µ or M), the coefficient of variation (s or S), and skewness (l or L) and to smooth the centile curves for IPFT indices against age, utilizing the method described by Cole TJ, et al. [18]. The goodness of fit for the model used was tested by the Q curve [18].

A total of 3412 neonates born in our institute from August 2012 to May 2014 were screened for eligibility; 310 neonates fulfilled the inclusion criteria and were enrolled. The median (IQR) age at enrolment was 4th (3rd, 5th) postnatal day. A total of 281 children (90.6%) had completed 36 months of age by 31 March 2017; all IPFT (TBFVL, RTC and RVRTC) at this age were successfully performed in 225 infants (54.5% males) (Fig. 1). The mean (SD) birthweight and length were 2.6 (0.6) kg and 47.7 (6.6) cm, respectively. Other demographic characteristics of enrolled children are summarized in Table I. TBFVL, RTC, and RVRTC were successfully performed in 1705, 948, and 875 occasions (baseline and follow up visits), respectively. The 5th, 25th 50th, 75th, and 95th centile values of TBFVL and RTC as per the age, sex, mean weight and length/height of the enrolled children are described in Table II and III. The same for RVRTC indices are described in Table IV.

TABLE I	Demographic Characteristics of Enrolled Infants (N=310)

TABLE II Normal Values (Percentiles) of Tidal Breathing Flow Volume Loop Indices

TABLE III Normal Values (Percentiles) of Rapid Thoracoabdominal Compression Indices

TABLE IV Normal Values of Raised Volume Rapid Thoracoabdominal Compression Indices

Tidal volume (VT) per unit body weight (VT/kg) increased significantly from birth to 36 months of age (P<0.001), which was more prominent in the first 12 months of age (P<0.001). The ratio of time to attained peak expiratory flow to expiratory time (tPTEF/tE) decreased significantly (P<0.001) from baseline till 18 months of age, then it began to increase (P<0.001) to achieve the baseline value at 36 months of age. VPTEF/VE also had decreasing trends from baseline to 18 months (P<0.001) then it gradually increased to achieve the baseline value by 36 months of age. The smoothed centile curves for VT and tPTEF/tE are given in Fig. 2 and 3. At baseline, there was no significant association of gender with IPFT indices. However, on follow up, VT was significantly more in boys at age of 6 months (P<0.001), 12 months (P<0.001), 18 months (P=0.02) and 36 months (P<0.001). The tPTEF/tE was significantly more in girls at 6 months (P=0.004) and 30 months (P=0.04) of age; for other ages it was similar in both sex. VPTEF/VE was significantly greater in girls at 6 months (P=0.004) and 30 months (P=0.02) of age. The tE/ttot was similar in both sexes at all ages.

Fig. 1 Enrolment and follow-up of the study cohort.

Fig.2 (a) Change in tidal volume (VT) with age in girls; (b) Change in tidal volume (VT) with age in boys.

Fig. 3 (a) Change in ratio of time to attained peak tidal expiratory flow to expiratory time (tPTEF/tE　) with age in girls; (b) Change in ratio of time to attained peak tidal expiratory flow to expiratory time (tPTEF/tE　) with age in boys.

Fig. 4 (a) Change in maximum expiratory flow at functional residual capacity (V¢max,FRC ) with age in girls; (b) Change in maximum expiratory flow at functional residual capacity (V¢max,FRC ) with age in boys.

In RTC, V¢maxFRC, V¢50, and V¢70 did not significantly differ from baseline to 36 months of age, while VPEF gradually increased from baseline to 36 months of age (P<0.001). The smoothed centile curves for V¢maxFRC are presented in Fig. 4. V¢max,FRC, V¢50, V¢70, and VPEF were similar in both sexes at all ages except at 30 months, where V¢70 (P=0.01) and VPEF (P<0.001) were significantly higher in boys. In RVRTC, there were minor increases in FVC, FEV0.5, FEF25-75, and MEF25 from baseline till 24 months of age; after that, there were significant increases till 36 months of age (P<0.01).

Fig. 5 (a) Change in forced expiratory volume in 0.5 sec (FEV0.5　) with age in girls; (b) Change in forced expiratory volume in 0.5 sec (FEV0.5　) with age in boys.

PEF initially decreased from baseline to 12 months of age (P<0.001) then remained constant till 24-months of age, increasing again till 36 months of age (P<0.001). FEV0.5/FVC was significantly decreased from baseline to 36-months of age (P=0.002), more during baseline to 18 months of age (P<0.001). The smoothed centile curves for FEV0.5 and FVC are represented in Fig. 5 and 6.

Fig.6 Change in forced vital capacity (FVC) with age.

RVRTC indices were similar in both sexes at all ages except at 36-months, where FEV, MEF25, FEF25-75, FEV0.5 and PEF, were significantly higher in boys (P=0.04, 0.01, 0.01, <0.001 and 0.02, respectively). FEV0.5/FVC remained similar in both sexes at all ages.

On multivariate analysis, length/height, body weight, gender, and age were significantly associated with VT. tPTEF/tE was significantly associated with gender, with values at any point being higher for females in comparison to males, after adjusting length/height and bodyweight.

In comparison to studies on Caucasian population, VT/kg was markedly lower at baseline in this cohort; however, it gradually increased with age, and by 12 months it became comparable to global values [19,20]. The rest of the TBFVL indices at baseline were comparable with other studies [19-21]. In this cohort, tPTEF/tE and VPTEF/VE were highest at baseline and then gradually decreased until 18 months of age. A study from Switzerland in 342 infants had also observed that tPTEF/tE gradually decreased in the first year of life [20]. However, a prospective birth cohort study from Taiwan did not observe any significant change in VT, tPTEF/tE and VPTEF/VE from 5 to 26 months of follow-up [7]. Furthermore, in the present study, there was no significant influence of gender at baseline IPFT indices; however, on follow up, many of these indices varied significantly with gender. A study from Norway did not find any significant influence of gender on the baseline tidal expiratory volume; however, they observed that tidal flow and flow ratio were significantly higher in males in comparison to female babies [21]. Another study from Taiwan also did not observe any sex-related difference in IPFT indices [7]. V¢max,FRC in this cohort remained similar throughout the follow-up period with no influence of gender. Studies from Taiwan [7], US [22], and a multicentric study from London, Indianapolis, and Boston [23] had observed that V¢max,FRC correlates significantly with the height of the children. The measurement of V¢max,FRC depends on accurate determination of FRC as a volume landmark, which is highly variable, especially in younger children, and this is a significant limitation in RTC measurement [24]. In this study, jacket pressure was kept fixed at 10 kPa while in other studies, it was used in the incremental range from 2-10 kPa [22,25]. Hence, it might be responsible for the deviation of our finding for V¢max,FRC with age from other studies.

In this cohort, most of the RVRTC indices viz., FVC, FEV0.5, FEF25-75, MEF25 and PEF increased minimally until 24 months of age, after that they increased dramatically with age. FEV0.5/FVC decreased from baseline until 36 months of age. There was no gender difference in any of the RVRTC indices except at 36 months where it was more significant in boys. A multicentric study from Indiana and Ohio [25] in children from 3 to 149 weeks had also observed that RVRTC indices were highly correlated with growth (height) of the child while FEV0.5/FVC decreased with increasing length. They also did not find any influence of gender on RVRTC indices except for FEF75, which was higher in girls [26]. Another study from London has documented that RVRTC indices increase with age [25].