This study was approved by our institutional research and ethical committee.

We carried out a retrospective study; the reference population comprised of healthy adults who were aged 18-70 years, who had undergone preventive health check ups in our hospital from January to December of 2011. Their medical history and clinical examination details were retrieved from the medical records. Out of the 2443 case sheets which were screened, five hundred (20.4% of the subjects) were selected, based on the inclusion and the exclusion as per the IFCC protocol [11].

18-70 years of age

Pathophysiological States - Renal failure, cardiac diseases, chronic respiratory diseases, liver diseases, malabsorption syndromes, malignancies and haematological disorders which included anaemias.

The institutional SOPs (Standard Operating Procedures) were followed for the sample collection and for conducting the tests.

All samples were collected between 7.00 am and 10 am. 3ml of whole blood was collected from the cubital vein with a vacutainer system into k3 EDTA tubes.

Tests: A complete haemogram was done by using a Sysmex KX-21 haematology analyzer within four hours of the blood collection. Dedicated reagents and standard methodologies were used. The 2-level quality controls were run every day and the analyzer was maintained according to the manufacturer’s instructions during the entire period of the study. The ESR estimation was done by using the Westergren method according to the ICSH protocol [12].

The data for the haematological parameters were collected and analyzed by using SPSS, version 16.0. The point estimate of the mean and the median with an interval estimate of 2.5 percentile and 97.5 percentile were provided as the reference values. The inferential statistics student t-test was done for the difference of all the parameters between the genders and the age at a 5% level of significance.

Among the 500 individuals, 248 (49.6%) were females and 252 (50.4%) were males. [Table/Fig-1] shows the age-gender distribution of the reference individuals. A majority of them (70%) were in the 25 to 55 years age group.

Frequency histograms were prepared for all the analytes and they showed the Gaussian distribution. [Table/Fig-2] and [Table/Fig-3] showed the Gaussian distribution of the RDW among the males and females in the histogram.

The means ± standard error and the study reference interval (2.5 – 97.5 percentile) for the males and the females are presented in [Table/Fig-4]. The reference intervals were calculated, based on the IFCC and the CLSI [13] guidelines. The males had a higher mean RBC count (5.10 X1012/L versus 4.39X1012/L), haemoglobin (14.43 g/dL versus 12.14 g/dL), haematocrit (41.63% versus 35.85%), MCH (28.74 pg versus 28.07 pg), MCHC (34.81% versus 34.19%) and PDW (13.07fL versus 11.96fL), whereas the females had a higher mean WBC count (8.45 X109/L versus 8.26 X109/L), MCV (81.77 fL versus 81.67 fL), Platelet count (261.60 X109/L versus 242.26 X109/L), RDW (13.72% versus 13.66%) and MPV (10.10 fL versus 9.71 fL) than their counterparts.

Statistically significant gender based differences in the means were observed for all the RBC parameters ( i.e. RBC count, haemoglobin, haematocrit, MCH and MCHC) except for MCV and RDW, for all the platelet parameters (Platelet count, PDW and MPV) and for ESR. The WBC (total count, differential lymphocytes%, and neutrophils %) did not show significant differences among the genders except for a differential mixed %.

Our study reference intervals were compared with the existing reference values [Table/Fig-5]. These values are currently in use in our laboratory and they were derived, based on the literature and the standard reference books [14-17]. The lower limits of the RBC count, haemoglobin, haematocrit, MCV and MCH were comparably lower; more so in females. Both the lower and the upper limits showed an increase with the RDW and the MPV values in both males and females. MCHC and the WBC parameters, the total count and the differential lymphocyte % showed a mild increase in the upper limits. An increase in the upper limit of ESR (with 30.2% males and 38.3% females as the outliers) was observed in comparison with the existing reference values. MCHC showed the maximum percentage of the study individuals (64.3% of the males and 44% of the females) outside the existing reference values, followed by MCV (63.1% of the males and 59.3% of the females).

[Table/Fig-1]:

Age-Gender distribution of Reference Individuals

[Table/Fig-2]:

Histogram of RDW for Males

[Table/Fig-3]:

Histogram of RDW for females

The reference individuals were further sub classified into more homogenous groups by stratifying them according to two intervals of age (<45 and >45years).

[Table/Fig-6] shows the mean values of the males and females when they were stratified according to the age. Statistically significant age related differences in the mean values of the RBC, haemoglobin, MCV, MCH, MCHC, platelets, PDW and ESR were observed in the males. However, among the females, haemoglobin, MPV and ESR showed statistically significant age related differences in the mean values.

To understand the effect of age and gender on the RDW, graphs were plotted between the mean value and the age [Table/Fig-7]. Females showed a gradual rise in the RDW values with age, but the RDW values of the males showed a decreasing trend.

We observed a high exclusion rate of about 80% while we conducted the study, based on the inclusion and the exclusion criteria. This was because of the high prevalence of clinical anaemia and Diabetes mellitus in our local population.

Statistically significant gender based differences were found for the following parameters and hence, separate reference intervals for the two genders should be considered: RBC count, haemoglobin, haematocrit, MCH, MCHC, Platelets, MPV, PDW and ESR.

We observed that when our reference intervals were compared with the existing reference ranges which are primarily derived from the western literature, there was a decrease in our study reference intervals for all the red cell parameters except for MCHC in both the genders. This was in agreement with the observations which were made by other Indian authors [18,19]. This decrease could be attributed to the prevalence of nutritional deficiencies and worm infestations in our population and this issue needs to be further addressed. The percentage of the reference individuals with values outside the current reference, ranged between 1-64%. The outliers were more for the RBC parameters and for ESR. This highlights the necessity of deriving region specific reference ranges.

The platelet count values did not show much differences from the existing reference values (5-6% outliers). This observation was in contrast to those of the previous studies from our region [20], who had reported a lower upper limit of the platelet count.

The total WBC count and the differential lymphocyte percentage and importantly, the ESR values were higher among our reference individuals. This could be due to the vulnerability of our population to infections. All the RBC parameters except RDW showed a statistically significant increase with age, particularly in males. This increase was not evident in females, though haemoglobin showed a significant change. Both of them showed an increase in the ESR values with age.

In our study, we observed an increase in both the limits of the RDW and the MPV for both the genders as compared to that in the literature [15,17]. The possibility of the prevalence of sub clinical nutritional anaemias would have had an impact on the RDW. At the same time, an increase in the risk of CVD among the local population cannot be under estimated [21] and its risk factors could have influenced both the parameters. This observation merits further studies.

Gender specific reference intervals are essential as there were statistically significant gender related differences in the RBC parameters, the platelet parameters and ESR. The reference intervals which were established by our study differed from the existing reference values.

A high increase in the upper limits of the ESR which was both age and gender related, shows that age as well as the gender based reference intervals needed to be considered for this parameter.

Though it is practically difficult to include age specific reference intervals in our haematology report for all the cases, it is always useful while studies are conducted where such parameters are used.

We derived the reference intervals for the newer parameters in our population as: RDW:

12.23%-15.36% in males and 12.3%-15.85% in females

We prefer gender specific reference intervals for all the three parameters.

We observed an increase in both the limits of the RDW and the MPV values in both males and females. This is a very valuable observation to define the cutoff points for both of them while evaluating for the risk of CVD and its associated mortality.

A population based study which correlates ESR with the other markers of inflammation might throw light on the reasons for the higher values of this parameter in our reference population.

A large cohort study is essential to precisely evaluate the age related changes in the RDW, MPV and the PDW.

An analytical study on the RDW and the MPV changes in the individuals who are at a risk for cardiovascular diseases may also be contributory.