Ayesha Molla ( Department of Pathology, The Aga Khan University Hospital, Karachi. )
Mohammad Khurshid ( Departments of Pathology, The Aga Khan University Hospital, Karachi. )
William T. Manser ( Department of Pathology, Baqai Medical College, Karachi. )
Rukhsana Lalani ( Department of Pathology, The Aga Than University Hospital, Karachi. )
Anis Alam ( Department of Pathology, The Aga Than University Hospital, Karachi. )
Zubaida Mohammad ( Department of Pathology, The Aga Than University Hospital, Karachi. )
Seven hundred and eighty six apparently healthy males (418) and females (368) aged 0-69 years were randomly selected for estimation of reference ranges of 24 serum analytes at the clinical chemistry laboratory of The Ago Khon University Hospital (AKUH). Of the total study samples, 56% (439/786) were in the poediatric age group (0-14 years) and 44% (347/786) in the adult (1 5_60 years) group. Beckman Astra Ideal Autoanalyzer was used for all the estimations. Moon and standard deviations (SD) were calculated for each of the age groups. Reference ranges were calculated following standard methods of the International Federation of Clinical Chemistry (IFCC) and International Committee far Standardization in Haematology (ICSH) (JPMA 43:113, 1993).
In a Clinical chemistry laboratory, established normal reference range of serum biochemical analytes is an essential information needed for comparison of the estimated values obtained from the patients. The validity of a reference range depends on various factors, e.g., subject, age, sex, methodologies, instruments used and overall laboratory environment. Previously some of the studies in Pakistan have reported normal ranges for some of the analytes1,2 estimated from a limited number of apparently healthy population. The objective of the present study is to estimate and establish a complete set of widely used reference range values of the serum analytes following a standardized criteria of IFCC and ICSH3-6.
Material and Methods
Seven hundred and eighty six healthy male and female volunteers aged between 0 _60 years were selected for the study. The population was stratified into two groups, e.g., pacdiatric aged between 0-14 years and adults aged between 15 _60 years. The adults were mostly employees, medical students, friends and relatives of the employees of AKUH. The paediatric group was selected mainly from a local private school and from the volunteering children of the employees of AKUFI. A questionnaire form including questions on dietary habits, recent and past history of illness, current use of medication and ethnic origin were asked for revealing any obvious reason (diabetes or liver disease) for not including their estimated results in the final data analysis. After an overnight fast, 10 ml of blood samples were collected from the selected subjects in neutral glass tubes. Serum was separated within an hour and all the analytes were estimated within 12 hours of collection. Beckman Astra 24 channel autoanalyzer (Beckman Instrument Inc., USA) was used for estimation of 24 biochemical analytes: glucose, cholesterol, triglycerides, BUN, creatinine, sodium, potassium, chloride, bicarbonate, direct and total bilirubin, gamma glutamyl transferase (GGT), glutamate pyruvate transaminase (SGPT), alkaline phosphatase, total protein, albumin, globulin, glutamate- oxaloacetate, transaminase (SGOT), lactate dehydrogenase (LDH), creatinine phosphokinase (CPK), calcium, phosphorus, amylase and uric acid. Regular maintenance of the instruments is done daily, weekly, bi-weekly, monthly and yearly. Equipment calibration for each of the chemistry is done 8 hourly, using standardized calibrator samples supplied by Beckman Instruments. Internal quality control samples obtained from commercially prepared serum (Beckman) with three levels (low, medium and high) of concentrations of each constituent are used four times daily. In between hours, home made standardized pooled serum samples are also used as internal quality control sample. External quality control samples supplied by Wellcome External Quality Assessment Scheme, UK7 are run twice monthly. All quality control results are scrutinized carefully before releasing patient’s results. Reporting of results are carried out according to the regulations for laboratory certification by the College of American pathologists8.
Means and standard deviations were calculated for each of the male and female groups. From any set of results, values above or below 2SD were excluded as outliers. Means and standard deviation were calculated for the second time following which lower and higher ranges were calculated by using the formula mean ± 2SD. This formula is only valid for those analytes, the distribution of which are of the Gaussian type. For the skewed pattern of distribution, following recommendation of IFCC, a non-pancreatic test was performed3-6 to set lower and higher range values.
1. Rehman, A and Naqvi, S.A.J. correlation between dietary protein intake, serum protein, blood urea nitrogen and serum creatinine level in apparently healthy males and females.). Pak. Med. Assoc., 1979;29:212-15.
2. lbrshim, K., Zubcri, S.J. and Hasnain, S.N. Serum alkaline phosphatase levels in apparently healthy subjects residing in Karachi. Pakistani. Med. Res., 1981;20:105-9.
3. Grasbeek, R., Siest, G., Wilding, P. Provisional recommendation on the theory of referencevalues. Part 1. The concept of referencevalues. din. chem., 1979;25:1506-8.
4. Patitclere, C., Wilding, P. The theory of referencevalues. Part 2. Selection of individuals for the production of reference values. J.Clin. chem. din. Biochem.. 1984;22:203-8.
5. Solberg, H.E. The theory of reference values. Part 5. Statistical treatment of collected reference values. Determination of reference limits. J.dlin.dhem.Clin.Biochem., 1983;21:749-60.
6. DybKaer, It The theory of reference values. Part 6. Presentation of observed values related to referencevalues. J.Clin.Chem.Clin. Biochem., 1982;201:841-45.
7. Murex Diagnostics Limited. Welleome External Quality Assessment Programmes. Central Road, Temple Hill, Dartford, England DAI 5LR.
8. Survey Manuals. An inter-laboratory comparison programme 1992, College of American Pathologists. 323 Waukegan Road, North field, Illinois 60093-2750, USA, 1992, pp. 8-17.
9. Normal Laborstosyvalues. N. EngI. 3. Med., 1986;314:35-41.
10. Reference ranges. Fundamentals of clinical chemistry. 3rd ed., Editor N.W. Tietz, Philadelphia, W.B. Saunders Company, 1987; pp.944.65.
11. Molla, A., Manser, W.W.T., Lalani, It, Badruddin, S.H., Mohammad, Land Khurshid, M. Blood lipids in a healthyKarachi population.). Trop. Med. Hyg., 1990;93:295-99.
12. Molla, A., Khurshid, M., Lalani, It, Manser, W.W.T. and Alam, A Serum alkaline phosphataae inapparently healthyKarachi population. ).Pak. Med. Assoc., 1990;40:182-84.
Reference range values in clinical chemistry may be defined as the result of a quantitative analysis of a group of healthy individuals selected at random according to clearly specified criteria. It was necessary to establish a reference range value for clinical chemistry which will be essentially applicable in Pakistani population. Using an internationally recognized and well calibrated equipment, e.g., Astra Autoanalyzer, we have attempted to develop our own values while vigorous quality control procedures were employed for obtaining precise and accurate results. For comparative analysis, the most up to date western reference range values used by the Massachusetts General Hospital9 and by International Federation of Clinical Chemistry10 are provided in Table W. In this study, we have stratified our population into two main groups providing separate reference range values for both paediatric males, females (Table II) and adult males, females (Table III) respectively. However, while comparing results between Pakistani and western adults, we obtained comparable results for most of the analytes except cholesterol, triglyceride, alkaline phosphate, lactate dehydrogenase, creatinine phosphatase and amylase values. Reference range values for these analytes are comparatively higher in our population than those of the western population as shown in Table IV. There could be various possible explanations for the discrepancy between these two sets of results obtained from two culturally different populations. Our results on cholesterol, triglyceride and alkaline phosphatase were critically analyzed and reported previously11,12.