Comparative study of hemoglobin estimated by Drabkin's and Sahli's methods.P Balasubramaniam, A Malathi
Department of Physiology, L.T.M.M. College, Sion, Bombay.
Hemoglobin was estimated by Sahli's and Drabkin's method in samples collected by finger prick and venepuncture. Statistical analysis revealed a significant difference in the hemoglobin values obtained by the two methods (P less than .01 & P less than .05 respectively). Sahli's method carried out by two different groups of workers in samples of blood collected by venepuncture showed statistically significant difference (P less than .001). This could be due to the inbuilt errors of Sahli's method including a subjective bias due to visual comparison. When same samples were subjected to Drabkin's method by the same group of workers no significant difference in the results were obtained (P less than .05). This is due to reliability of the Drabkins method. The aim of this comparative study of hemoglobin estimation by Sahli's and Drabkin's methods is to emphasize the sensitivity and reliability of Drabkin's over Sahli's and utility of Drabkin's method in undergraduate teaching schedule.
Keywords: Blood Specimen Collection, Comparative Study, Hemoglobinometry, methods,Human, Methemoglobin, analogs &derivatives,analysis,Reference Values,
Sahli's method of haemoglobin estimation has been used since long. In spite of the availability of newer techniques, which give more reliable and accurate result, this method is still in vogue. The International Nutritional Anaemia Consultive Group (INACG) and International Committee for Standarization in Hematology (ICSH) have recommended the Drabkins as the method of choice and have suggested that all other methods should be adjusted to be comparable to this method. For anaemia prevalence studies in India, institutes like NIN, Hyderabad have adopted modifications of Drabkin's method.
Hence an attempt has been made to establish the efficacy of Drabkin's method by comparison of Sahli's with Drabkin's method of haemoglobin estimation.
Haemoglobin was estimated by Sahliís and Drabkin's cyanmethemoglobin methods. Standardised haemoglobin pipettes were used in both methods. Drabkin's solution containing potassium ferricyanide,
potassium cyanide and sodium bicarbonate was prepared. 20 microliter of blood was added to 5ml of Drabkin's solution. Readings were taken at 530 nm in a spectrophotometer. Haemoglobin values were calculated from a haemoglobin curve prepared using haemoglobin standard (Chemkit).
Eighty three samples of blood collected by finger prick method and 132 samples by venepuncture were subjected to haemoglobin estimation by Sahli's and Drabkins cyanmethemoglobin methods, Haemoglobin values obtained by the two methods were compared. Ninety-two samples collected by venepuncture were subjected to haemoglobin estimation by Sahli's method by two different groups of workers and the haemoglobin values compared. The above two groups of workers also determined haemoglobin by Drabkin's method in 92 samples of blood collected by venepuncture. The haemoglobin values obtained by Drabkin's method by the two groups were compared.
Comparison of haemoglobin estimated by Sahli's and Drabkins methods by fingerprick and venepuncture method of collection showed a highly significant difference [Table - 1].
Sahli's method of estimation in the samples by two different groups of workers also showed a significant difference [Table - 2].
However, Drabkins method of haemoglobin by the two different workers did not show any significant difference [Table - 3].
A number of workers have compared haemoglobin estimated by different methods in order to establish the efficacy and reliability of the methods,. Some of the methods of haemoglobin estimation may have an error of + / -20% or more which when compounded with poor technique makes the method highly unreliable.
The principle in Sahli's method involves the conversion of haemoglobin to acid hematin and comparing visually the colour developed with that of standard tinted glass. Haemoglobin value is directly read from the
graduated haemoglobin tube. The methodology has following inbuilt disadvantages viz: i) subjective visual colour comparison, ii) need for accurate pipetting of 20 microliter of blood, iii) estimation of only acid hematin formed, iv) fading of comparator on prolonged use and v) poor sensitivity and reliability.
In the Drabkin's method of haemoglobin estimation haemoglobin is oxidised to methemoglobin by potassium ferricyanide, which reacts with cyanide ions of potassium cyanide to form cyanmethemoglobin. The haemoglobin is estimated with the help of cyanmethemoglobin curve. The advantages of this method are i) error due to subjective visual matching is avoided as spectrophotometer is used and hence reading is precise and reliable, ii) measures all forms of haemoglobin except sulphaemoglobin. iii) single step procedure using single reagent. iv) cyanmethemoglobin formed produces broad absorbent band at 530 rim v) good stable haemoglobin standards are available.
Analysis of the results of haemoglobin estimated by Sahli's and Drabkin's methods in samples of blood collected by fingerprick and venepuncture showed statistically significant difference. This could be due to the inbuilt errors of Sahli's method of haemoglobin estimation. The subjective error in colour comparison in Sahli's method is proved by the fact that readings repeated in the same samples but by two different groups of workers were significantly different. On the other hand comparison of haemoglobin estimated by Drabkin's method in the same samples by the two groups of workers were not statistically significant stressing thereby the reliability of Drabkin's method.
Thus the Drabkin's method of haemoglobin estimation cab be used in undergraduate teaching schedule as well as in anaemia detection and management in hospitals replacing Sahli's method. Since the cut off point decides the requirement of therapy, the Drabkin's method being more sensitive is the method of choice. Also, for research purposes where small variations in haemoglobin values are to be detected, the Drabkin's method would be ideal.
[Table - 1], [Table - 2], [Table - 3]