Efficacy of Dichlorophenolindophenol (DCIP) as Screening Test for Hb E: Revisited

Abstract

Hb E-beta thalassemia is a major public health problem in West Bengal, India and is the predominant symptom producing thalassemia in this part of the country. To search for an easy, reliable and cost effective screening method for HbE that can be used at the community level where more sophisticated methods are not readily available. And the DCIP test was performed for the purpose. Blood samples of 425 asymptomatic family members from 80 diagnosed cases of HbE beta Thalassemia patients were tested for Hb, RBC indices, DCIP test, HPLC, and in discordant cases confirmed by DNA mutation analysis. The present study shows DCIP screening test to have a sensitivity, specificity, positive predictive value and negative predictive value of 96.39%, 97.43%, 96.39% and 97.43% respectively. It also shows a false positive rate and false negative rate in 2.56% and 4.6% cases respectively. The advantage with DCIP over HPLC is that it can be easily performed at the community level by a person with minimum technical skill, few samples (even a single sample) can be tested at time, at a low cost.

Introduction

Haemoglobin E (HbE) is a common variant hemoglobin caused by point mutation in 26th position of β-chain that activates cryptic splice site at the 25th codon resulting in qualitative and quantitative defect of beta-globin chain [1]. HbE is the most frequent β-thalassemic hemoglobinopathy in Southeast Asia and parts of the Indian sub-continent [2]. It has a significant prevalence in North East India, Bangladesh, Indonesia and Sri Lanka. In India, in some foci in the North East the prevalence reaches 80%. The prevalence rate of HbE in Kolkata is as high as 22% [1].

Hemoglobin E in homozygous and heterozygous state is asymptomatic. But, if they get married with beta-thalassemia carriers, there is a chance of having Hb E-beta thalassemia offsprings, which is a form of symptomatic thalassemia. To prevent this, carrier detection by a cost-effective and robust method is important.

HPLC is used as gold standard to diagnose thalassemia and other hemoglobinopathies. But, it is costlier, large number of samples are needed at a time to make it cheaper, needs expertise skilled personnel and a well equipped laboratory set up. For the reasons mentioned above, it is not readily available at the community level.

Dichlorophenolindolphenol (DCIP) test is used to detect the presence of Hb E which is easily and quickly oxidized by DCIP reagent [3, 4]. In search of a screening test which is much cheaper, easy to perform, requires less expertise and can be done at the community level; we performed DCIP test.

Materials and Methods

The work conducted over a period of 5 years (January, 2012 to December, 2016) in asymptomatic family members of E-beta thalassemia patients attending thalassemia OPD and Thalassemia day care unit in the Department of Hematology, NRS Medical College, Kolkata. A total of 425 asymptomatic family members of 80 (eighty) E-beta thalassemia patients attending the thalassemia OPD and Day care center were studied. The parameters studied include detailed history and clinical examination, personal history of blood (RBC) transfusion and history of transfusion dependant anemia in the family; history of consanguineous marriage was noted carefully. The study excluded family members having symptomatic anemia, jaundice and/or hepato-splenomegaly. Samples of EDTA anticoagulated whole blood tested for complete hemogram including RBC indices, DCIP (dichloro-phenolindophenol) test and HPLC (high performance liquid chromatography). The results obtained from DCIP (dichloro-phenolindophenol) test were evaluated and compared with HPLC data. The cases showing discordance by the above mentioned methods were finally diagnosed and confirmed by DNA mutation analysis. ARMS-PCR (amplification refractory mutation system—polymerase chain reaction) technique was used for mutation analysis. Thus, effectiveness of DCIP (a screening test) e.g. sensitivity, specificity, positive predictive value and negative predictive value were compared with that of other costlier test e.g. HPLC.

Complete Hemogram

Done by automated cell counters using Sysmex KX-21 automated hematology analyzer. Cell counter parameters studied included hemoglobin (Hb%), hematocrit (Hct), Red blood cell (RBC) count and RBC indices e.g. MCV (mean corpuscular volume), MCH (mean corpuscular hemoglobin), MCHC (mean corpuscular hemoglobin concentration) and RDW-CV (red cell distribution width- coefficient of variation). Peripheral blood smears were examined to look for RBC morphology and also for presence of nucleated red blood cells (NRBCs), abnormal cells, if any.

DCIP (Dichloro-Phenolindophenol) Test (Also Known as DCPIP Test) [5]

2,6-Dichlorophenol-indophenol (DCIP) is a blue chemical compound used as a redox dye. HbE and other unstable haemoglobin molecules (such as HbH) are precipitated when exposed to this dye at 37 °C. Precipitated haemoglobin (if any) visualized by the naked eye at the bottom of the tube. In homozygous HbE, heavy sediment is formed; in HbE trait and Hb E/β thalassemia, the precipitation of Hb E produces a cloudy or an evenly distributed particulate appearance.

HPLC (high performance liquid chromatography)

Done by BIO-RAD ‘beta thalassemia short Programme VARIANT’ instrument. Typical ‘retention times’ observed were noted carefully and the hemoglobins are assigned [2] to retention time ‘windows’ which are designated as F, P₂, P₃, A, A₂, S, C and D.

ARMS-PCR (Amplification Refractory Mutation System–Polymerase Chain Reaction)

DNA was extracted from peripheral blood leucocytes by Phenol–chloroform method [6]. ARMS-PCR technique [7] was applied to confirm the presence of HbE mutation. Primers used to detect the presence of Hb E and beta thalassemia mutations (commonly found in this part of the country) [8] are shown in Table 1.

Table 1.

Following primers were used to detect the presence of Hb E and beta thalassemia mutations

Mutation	Primer sequence 5′ → 3′	Second primer	Fragment size (bp)
Codon26 (G → A) HbE	N- TAA CCT TGA TAC CA ACCT GCC CAG GGC GTC M- TAA CCT TGA TAC CA ACCT GCC CAG GGC GTT	C1 C1	236 236
IVS1-5 (G → C)	N- CTC CTT AAA CCT GTC TTG TAA CCT TGT TAC M- CTC CTT AAA CCT GTC TTG TAA CCT TGT TAG	C1 C1	285 285
IVS1-1 (G → T)	N- GAT GAA GTT GGT GGT GAG GCC CTG GGT AGG M- TTA AAC CTG TCT TGT AAC CTT GAT ACG AAA	C2 C1	454 281
Codon41/42 (-CTTT)	N- GAG TGG ACA GAT CCC CAA AGG ACT CAA AGA M- GAG TGG ACA GAT CCC CAA AGG ACT CAA CCT	C1 C1	443 439
Codon15 (G → A)	N- TGA GGA GAA GTC TGC CGT TAC TGC CCA GTG M- TGA GGA GAA GTC TGC CGT TAC TGC CCA GTA	C2 C2	500 500
Codon30 (G → C)	N- TAA ACC TGT CTT GTA ACC TTG ATA CCT ACC M- TAA ACC TGT CTT GTA ACC TTG ATA CCT ACG	C1 C1	280 280
Codon8/9 (+ G)	N- CCT TGC CCC ACA GGG CAG TAA CGG CAC ACT M- CCT TGC CCC ACA GGG CAG TAA CGG CAC ACC	C1 C1	214 215

C1 and C2 are common primes used in PCR reaction to check for amplification. All the primers used for ARMS are HPLC purified. Common primer 1 (C1):- 5′ ACC TCA CCC TGT GGA GCC AC3′. Common primer 2 (C2):- 5′ CCC CTT CCT ATG ACA TGA ACT TAA 3′

Results

A total of 425 asymptomatic family members from 80 (eighty) diagnosed cases of HbE-beta thalassemia patients included in the study. A total of 15 (fifteen) members from those families were never available for study because 9 (nine) members were elderly (unable to attend) and others (6 members) didn’t turn up in spite of every efforts. Thus, including all, average asymptomatic member per family was calculated as: (425 + 15)/80 = 440/80 = 5.5 asymptomatic members per family (range 2–16). Out of 425 cases, there were 198 (46.59%) male and 227 (53.41%) females. Median age was 23.95 years (range 1–75). Highest number of screened subjects (59.53%) belonged to the age group of 10–40 years.

In hemogram, MCH value < 27 pg was found in 256 (60.23%) cases and MCV < 80 fl seen in 241 (56.7%) cases. Smears examined for RBC morphology and compared with the RBC indices and finally with HPLC reports. Results of mean values of hemogram in normal, HbE trait and beta thalassemia trait shown in Fig. 1. Mean hemoglobin level, MCV, MCH, MCHC and RDW values of HbE/HbEE (heterozygous and homozygous states of Hemoglobin E) cases were in between that of normal subjects and beta thalassemia traits. The mean RBC count in Hb E/EE cases was much higher in comparison to normal and beta thalassemia trait cases. Peripheral blood smear examined carefully for RBC morphology showed microcytic, hypochromic RBC’s in beta thalassemia trait and HbE/HbEE cases. Out of 425 samples tested for DCIP (Fig. 2), 152 (32.76%) samples were positive for the test and 273 (67.24%) samples showed negative results. In majority of the positive cases there was a cloudy particulate appearance. In 4 cases, heavy sediment was formed at the bottom of the test tube. The results of HPLC reports are summarized in Fig. 3 showing comparison of mean values of HPLC in normal, HbE trait, beta thalassemia trait and HbEE cases. In 137 (32.23%) cases HPLC studies showed normal chromatographic pattern; 154 (36.24%) and 134 (31.53%) cases showed chromatographic pattern suggestive of HbE/HbEE and beta thalassemia traits respectively. Thus, HPLC study of family members showed a ratio of normal: HbE: beta thalassemia trait = 137: 154: 134 = 1.02: 1.15: 1.

Fig. 1.

comparison of mean values of hemogram in normal, HbE trait and Hb beta trait

Fig. 2.

tube at the center is blank, other two tubes showing precipitate at the bottom indicating positive result with DCIP

Fig. 3.

Comparison of mean values of HPLC in normal, Hb beta trait, HbE trait and HbEE

The test results of DCIP screening test and HPLC were now compared and summarized in Table 2. Out of 425 blood samples tested with both the procedures, in 14 (3.29%) cases there were discordance/discrepancy in results. As compared to HPLC, DCIP showed false positive and false negative results in 7 (1.65%) cases and 7 cases (1.65%) respectively.

Table 2.

Summary of results of DCIP, HPLC and ARMS-PCR result

Case no. (sample no.)	DCIP test	HPLC report	Inference (DCIP compared to HPLC)	ARMS-PCR result	Final diagnosis
19	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Heterozygous for HbE mutation
31	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Heterozygous for HbE mutation
36	(+) ve	Beta trait	False + ve	Codon8/9 (+ G)	Heterozygous for beta mutation
84	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Heterozygous for HbE mutation
86	(+) ve	Beta trait	False + ve	IVS1-5 (G → C)	Heterozygous for beta mutation
131	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Homozygous EE
206	(+) ve	Normal	False + ve	No mutation detected	Normal
269	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	heterozygous for HbE mutation
277	(+) ve	Normal	False + ve	No mutation detected	Normal
298	(+) ve	Beta trait	False + ve	Codon8/9 (+G)	heterozygous for beta mutation
307	(+) ve	Normal	False + ve	No mutation detected	Normal
356	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Heterozygous for HbE mutation
395	(+) ve	Normal	False + ve	No mutation detected	Normal
401	(−) ve	E trait	False − ve	Codon26 (G → A) HbE	Heterozygous for HbE mutation

Blood samples of all those 14 cases showing discordance/discrepancy in results between DCIP screening test and HPLC test, processed for DNA analysis by ARMS-PCR. All the samples were first tested for HbE mutations; the mutation was detected in 7 cases (heterozygous-6 cases and homozygous-1 case) which gave false negative results in DCIP test as compared to HPLC. DNA samples showing no HbE mutation (7 cases) were now subjected to studies for the common beta thalassemia mutations; in 3 cases there was beta thalassemia mutation and in rest 4 cases no mutation was detected (hence, reported as normal). Summary of tested results of DCIP, HPLC and ARMS-PCR result are shown in Table 2.

Validation of DCIP Screening Test Results

Among 425 samples tested, there was discordance in 14 (3.29%) cases (false positive-7 cases and false negative-7 cases) as compared to HPLC and ARMS-PCR results. Thus, the present study with DCIP test shows sensitivity of 96.39% and specificity of 97.43%. In assessing the diagnostic power of the (DCIP screening) test, it showed positive predictive value, negative predictive value, percentage of false positives and percentage of false negatives of 96.39%, 97.43%, 2.56% and 4.6% respectively.

Discussion

Haemoglobin E, a hereditary abnormality of human hemoglobin was first described by Chernoff et al. [8]. Independently in the same year it was also described by Itano et al. [9] as fourth abnormal hemoglobin. Since its classic description by Chernoff et al. [8], it has been found to be an important public health problem in the Indian subcontinent and Southeast Asia.

The first case of Hb E/β-thalassemia in India was reported by Chatterjea et al. [10] from Calcutta. Thalassemia and other hemoglobinopathies is a major public health problem in West Bengal, India; Hb E-beta thalassemia is the predominant symptom producing thalassemia in this part of the country. The prevalence rate of HbE in this state [11] is 4.1% with reported very high prevalence of 22% in Kolkata [1], the capital city of West Bengal. These established facts are the major driving force in searching for a suitable and ideal screening test to be performed at the community level.

Hemoglobin E in homozygous and heterozygous state is asymptomatic. But, if they get married with beta thalassemia carriers, there is a chance of having Hb E-beta thalassemia offsprings, which may give rise to a form of symptomatic thalassemia. To prevent this, carrier detection by a cost-effective and robust method is necessary.

At present in India and also in West Bengal, HPLC is used as a screening test as well as confirmatory test for diagnosis of HbE and also other hemoglobinopathies and thalassemias. Though it gives reliable and reproducible results, but it is costlier, needs expertise personnel and well equipped laboratory set up and thus not available for population screening at the community level. The present study was a blinded study as the blood samples were first tested by automated cell counter, DCIP test; then subjected to HPLC test. The results coming out of automated cell counter and DCIP test were then corroborated with HPLC test results. In the present study, we evaluated the effectiveness of DCIP screening test for HbE in corroboration with the HPLC reports and any discordance thus coming out between the two methods were finally confirmed by DNA mutation analysis by ARMS-PCR method. In the present study, total of 425 asymptomatic members from 80 diagnosed cases of E-beta thalassemia patients were included in the study. Fifteen members from those families were never available for study because 9 (nine) members were elderly (unable to attend) and others (6 members) didn’t turn up in spite of every efforts.

Mean Hb A₂ + E level in Hb E and HbEE were 28.3% and 90% respectively which was at par (29.3%) with studies by Sanchaisuriya et al. [12] done in Thai population. Mean Hb A₂ + E level in heterozygous state was 28.3% which is slightly less in comparison to other reports from south India (30.1%) and western countries (30%) [13, 14]. This low level of Hb A₂ + E in eastern India and Southeast Asia are explained by possible concurrent presence of alpha-thalassemia mutations in this part of the globe which can only be effectively diagnosed by mutation analysis for alpha thalassemia [15].

HbE was detected in 154 (53.1%) cases among the family members and beta thalassemia trait was detected in 134 (46.9%); thus, the present study reports a higher prevalence rate of HbE, possibly because of biasness in selecting the study population including only the family members of diagnosed HbE beta thalassemia patients. This is in contrast to other studies [2, 11], where they performed the study in general population. Study conducted by Balgir [16] on ‘the genetic epidemiology of the three predominant abnormal hemoglobins in India’ showed that HbE is mostly restricted to the north-east part of India i.e. West Bengal, Assam, Nagaland, Manipur, Tripura and Meghalaya with an average allele frequency of 10.9% with occasional case reports from other part of the country.

Mean Hb % values of 11.8 gm/dl, 10.74 gm/dl and 12.3 were found in HbE/EE, beta thalassemia trait and normal cases respectively (Fig. 1). Mean RBC count was much higher in both conditions (4.69 × 10¹²/µl and 5.29 × 10¹²/µl) in comparison to normal cases (4.11 × 10¹²/µl) and mean MCH of 25.35 pg and 20.7 pg was lower than the normal cases (29.1 pg). MCH value < 27 pg was found in 256 (60.23%) cases and MCV < 80 fl seen in 241 (56.7%) cases. Study by Sanchaisuriya et al. [12], considering MCH value < 27 pg and MCV < 80 fl for screening, showed a sensitivity, specificity, positive predictive value and negative predictive value of 78.9%, 79%, 72.6%, 84.15% and 72.0%, 84.3%, 76.4% and 81% respectively. Many of the HbE heterozygotes shown normal RBC indices; therefore gave high false negative results; and they showed that, sensitivity of MCV and MCH screening protocol was improved to 100% when it was combined with DCIP test.

In this study, out of 425 samples tested for DCIP, 152 (32.76%) samples were positive for the test and 273 (67.24%) samples showed negative results. The samples then subjected to HPLC test; showed results as follows: normal, beta thalassemia trait, HbE and HbEE in 137 cases, 134 cases, 150 cases and 4 cases respectively. Thus, DCIP test results when corroborated with HPLC test results (Table 2) gave a false positive of 7 cases and false negative of 7 cases respectively. Tyagi et al. [17] from All India Institute of Medical Sciences, New Delhi concluded that, HPLC being an automated instrument is highly sensitive and specific, has high resolution and helps in quantification of various hemoglobins. However, in a developing country like India where economical factors play a major role in planning for management of patients, the role of HPLC is limited.

The present study shows DCIP screening test to have a sensitivity, specificity, positive predictive value and negative predictive value of 96.39%, 97.43%, 96.39% and 97.43% respectively. It also shows percentage of false positives and percentage of false negatives in 2.56% and 4.6% cases respectively. As shown in Table 3, many other studies also reported the effectiveness of using DCIP and found that this test has a sensitivity of 94.4–100%, specificity of 69.8–98.2%, positive predictive value of 75.0–86.9%, and negative predictive value of 98.1–100%; can be used as an effective screening test for detection of Hb E [4, 12, 18–20]. The present study in comparison to others resulted in a slightly higher value of false negative rate which may impart a negative impact on the success of screening programme that may possibly be overcame by performing the study in a larger population as shown by study [19] done with a large number of cases.

Table 3.

Comparison of different studies regarding effectiveness of DCIP test

Study group	Sensitivity (%)	Specificity (%)	Positive predictive value (%)	Negative predictive value (%)	False positive rates (%)	False negative rates (%)
Present study	96.39	97.43	96.39	97.43	2.56	4.6
Prayongratana et al. [19]	97.16	98.93	99.42	95.19	1.07	2.66
Sanchaisuriya et al. [12]	100	76.2	74	100	NA	NA
Fuchareon et al. [4]	100	69.8	77.2	100	NA	NA
Chapple et al. [18]	100	92	85.7	100	NA	NA
Jaiwang et al. [20]	100	100	NA	NA	NA	NA

NA not available

The criteria for screening are based on two considerations: the disease to be screened, and the test to be applied [21]. The disease to be screened should fulfill the following criteria: (1) the condition sought an important public health problem, (2) facilities available for confirmation of diagnosis, and (3) early detection reduces morbidity, mortality and disease burden in the community. The test also must satisfy the criteria of acceptability, repeatability and validity, besides others such as simplicity, safety, rapidity and cost [22]. Performance of a screening test is measured by it’s predictive value which reflects the diagnostic power of the test. The present study showed a positive predictive value, Negative predictive value of 96.39% and 97.43% respectively. As evident, DCIP test may be accepted an effective screening test considering simplicity, rapidity, validity and cost.

However, there were certain limitations in the study; (a) small sample size: diagnostic power of the test e.g. predictive value can possibly be improved further by increasing sample size working with a larger population. (b) Biasness: the present study was conducted including a biased population of asymptomatic family members of the HbE beta thalassemia patients which can be minimized by performing the test in general population having high prevalence of HbE, (c) DNA mutation analysis by ARMS-PCR was performed in those cases only showing discordance between DCIP and HPLC for HbE and common beta thalassemia mutations reported from this part of the country.

The advantage with DCIP over HPLC is that it can be easily performed at the community level by a person with minimum technical skill, few samples (even a single sample) can be tested at time at a low cost. DCIP solution can be stored at 4 °C for a long time in a refrigerator, which is freely available nowadays even in the community health centers. In contrast, HPLC is needs highly skilled and dedicated technical personnel, a well organized laboratory set up and the running cost is very high, if done in a small number of samples.

Conclusion

DCIP test is a simple, easy to perform and cost effective method for detecting HbE; can be used in rural areas in a population with high prevalence with access to limited health care facilities.

Funding

None.

Compliance with Ethical Standards

Conflict of interest

The authors declare no potential conflicts of interest.

Ethical Approval

The study was approved by the Institutional Ethical Committee (IEC).

Patient Consent

Written consent taken from each patient and/or their legal guardians at the entry into the study.