On the nature of Romanowsky dyes and the Romanowsky-Giemsa effect

Abstract

Summary This paper reviews the nature of Romanowsky staining and the relationship between Romanowsky dyes and the Romanowsky‐Giemsa effect (RGE). On blood and bone marrow smears the RGE is characterized by a purple colouration of nuclei and neutrophil granules. The nuclear purple contrasts strongly with the blue cytoplasmic staining of cells rich in RNA. Requirement for the occurrence of RGE are: 1 A cationic dye: The best dye is azure B and, though azure A gives the nuclear purple colour, the cytoplasmic blue is inferior. No other cationic dye such as methylene blue is suitable. 2 An anionic dye: Most commonly eosin Y is used, but it can be replaced by the erythrosins. Full halogenation of the fluorescein (four atoms of bromine or iodine) is not necessary. Phloxine and rose bengal are unsuitable. 3 An appropriate substrate: These are proteins with acidic side groups or proteins bound to a polyanion. For interaction with the dyes substrates must provide a suitable three‐dimensional network which is why the RGE is not obtained in solutions. A tentative theory of RGE is advanced and briefly discussed.

A multitude of dyes but unsatisfactory staining: Physico-chemical profile and factors associated with staining quality of Romanowsky-type stains used for malaria microscopy in Plateau State, Nigeria

“…The higher the glycerol to methanol ratio, the poorer the performance of the stains. These findings tend to be in line with factors that have been previously demonstrated to influence performance of the Romanowsky-Giemsa stains (Marshall et al, 1975a, 1975c; Marshall, 1975a; Wittekind, 2008; Wittekind and Gehring, 1985). We have shown that minimum amounts of the RGE-relevant dyes are required to achieve the Romanowsky-Giemsa effect, which enables adequate cytoplasmic and nuclear colour differentiation in a stained malaria parasite.…”

“…In this current study, excellent staining of parasites was obtained with the reference stain (Marshall et al, 1978), which yielded 25.0% and 32.8% of Azure B and Eosin Y respectively with an attendant Azure B to Eosin Y ratio of 0.8:1.0 A similarly performing stain was a Giemsa sample from the ANDi Laboratory, Lagos, which gave a ratio of 1.1. These findings are at variance with previous studies that demonstrated satisfactory staining with Azure B to Eosin Y ratio ranging from 5:1 -10:1 (Wittekind, 1979). We, however, noted in this current study, a wide range of Azure B to Eosin Y ratio among the stain samples obtained from various laboratories and suppliers.…”

“…Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a;Friedrich K, Seiffert W, 1990;R. W. Horobin, 2011;WITTEKIND, 2008;Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and strength and direction of the association between each predictor and the likelihood of "satisfactory" staining outcome.…”

“…Ultraviolet spectroscopic and HPLC assays have revealed a high degree of variability of dye composition and the presence of RGE-irrelevant and RGE-antagonizing dyes in the stain samples analyzed in this study. Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a; Friedrich K, Seiffert W, 1990; R. W. Horobin, 2011; WITTEKIND, 2008; Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and Richard W. Horobin, 1985; Horobin and Walter, 1987a; Marshall et al, 1975a; Schulte, 1991).…”

A multitude of dyes but unsatisfactory staining: Physico-chemical profile and factors associated with staining quality of Romanowsky-type stains used for malaria microscopy in Plateau State, Nigeria

“…The higher the glycerol to methanol ratio, the poorer the performance of the stains. These findings tend to be in line with factors that have been previously demonstrated to influence performance of the Romanowsky-Giemsa stains (Marshall et al, 1975a, 1975c; Marshall, 1975a; Wittekind, 2008; Wittekind and Gehring, 1985). We have shown that minimum amounts of the RGE-relevant dyes are required to achieve the Romanowsky-Giemsa effect, which enables adequate cytoplasmic and nuclear colour differentiation in a stained malaria parasite.…”

“…In this current study, excellent staining of parasites was obtained with the reference stain (Marshall et al, 1978), which yielded 25.0% and 32.8% of Azure B and Eosin Y respectively with an attendant Azure B to Eosin Y ratio of 0.8:1.0 A similarly performing stain was a Giemsa sample from the ANDi Laboratory, Lagos, which gave a ratio of 1.1. These findings are at variance with previous studies that demonstrated satisfactory staining with Azure B to Eosin Y ratio ranging from 5:1 -10:1 (Wittekind, 1979). We, however, noted in this current study, a wide range of Azure B to Eosin Y ratio among the stain samples obtained from various laboratories and suppliers.…”

“…Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a;Friedrich K, Seiffert W, 1990;R. W. Horobin, 2011;WITTEKIND, 2008;Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and strength and direction of the association between each predictor and the likelihood of "satisfactory" staining outcome.…”

“…Ultraviolet spectroscopic and HPLC assays have revealed a high degree of variability of dye composition and the presence of RGE-irrelevant and RGE-antagonizing dyes in the stain samples analyzed in this study. Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a; Friedrich K, Seiffert W, 1990; R. W. Horobin, 2011; WITTEKIND, 2008; Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and Richard W. Horobin, 1985; Horobin and Walter, 1987a; Marshall et al, 1975a; Schulte, 1991).…”

Quantitative diagnostic algorithm using endocytoscopy for superficial nonampullary duodenal epithelial tumors

“…Unexpectedly, the correlation between the median number of DSMs per 1 mm 2 in ECS images and the number of nuclei in histology was weak. As described in a previous report, we think the reason for this is that DSMs possibly include not only nuclei but also cytoplasm 20 . Furthermore, a recent retrospective study showed that ECS diagnosis with methylene blue provided a high accuracy rate, and double staining using crystal violet was not recommended for the diagnosis of SNADETs 10 .…”

A multitude of dyes but unsatisfactory staining: Physico-chemical profile and factors associated with staining quality of Romanowsky-type stains used for malaria microscopy in Plateau State, Nigeria

“…The higher the glycerol to methanol ratio, the poorer the performance of the stains. These findings tend to be in line with factors that have been previously demonstrated to influence performance of the Romanowsky-Giemsa stains (Marshall et al, 1975a, 1975c; Marshall, 1975a; Wittekind, 2008; Wittekind and Gehring, 1985). We have shown that minimum amounts of the RGE-relevant dyes are required to achieve the Romanowsky-Giemsa effect, which enables adequate cytoplasmic and nuclear colour differentiation in a stained malaria parasite.…”

“…In this current study, excellent staining of parasites was obtained with the reference stain (Marshall et al, 1978), which yielded 25.0% and 32.8% of Azure B and Eosin Y respectively with an attendant Azure B to Eosin Y ratio of 0.8:1.0 A similarly performing stain was a Giemsa sample from the ANDi Laboratory, Lagos, which gave a ratio of 1.1. These findings are at variance with previous studies that demonstrated satisfactory staining with Azure B to Eosin Y ratio ranging from 5:1 -10:1 (Wittekind, 1979). We, however, noted in this current study, a wide range of Azure B to Eosin Y ratio among the stain samples obtained from various laboratories and suppliers.…”

“…Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a;Friedrich K, Seiffert W, 1990;R. W. Horobin, 2011;WITTEKIND, 2008;Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and strength and direction of the association between each predictor and the likelihood of "satisfactory" staining outcome.…”

“…Ultraviolet spectroscopic and HPLC assays have revealed a high degree of variability of dye composition and the presence of RGE-irrelevant and RGE-antagonizing dyes in the stain samples analyzed in this study. Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a; Friedrich K, Seiffert W, 1990; R. W. Horobin, 2011; WITTEKIND, 2008; Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and Richard W. Horobin, 1985; Horobin and Walter, 1987a; Marshall et al, 1975a; Schulte, 1991).…”

Quantitative diagnostic algorithm using endocytoscopy for superficial nonampullary duodenal epithelial tumors

“…Unexpectedly, the correlation between the median number of DSMs per 1 mm 2 in ECS images and the number of nuclei in histology was weak. As described in a previous report, we think the reason for this is that DSMs possibly include not only nuclei but also cytoplasm 20 . Furthermore, a recent retrospective study showed that ECS diagnosis with methylene blue provided a high accuracy rate, and double staining using crystal violet was not recommended for the diagnosis of SNADETs 10 .…”

A multitude of dyes but unsatisfactory staining: Physico-chemical profile and factors associated with staining quality of Romanowsky-type stains used for malaria microscopy in Plateau State, Nigeria

“…The higher the glycerol to methanol ratio, the poorer the performance of the stains. These findings tend to be in line with factors that have been previously demonstrated to influence performance of the Romanowsky-Giemsa stains (Marshall et al, 1975a, 1975c; Marshall, 1975a; Wittekind, 2008; Wittekind and Gehring, 1985). We have shown that minimum amounts of the RGE-relevant dyes are required to achieve the Romanowsky-Giemsa effect, which enables adequate cytoplasmic and nuclear colour differentiation in a stained malaria parasite.…”

“…In this current study, excellent staining of parasites was obtained with the reference stain (Marshall et al, 1978), which yielded 25.0% and 32.8% of Azure B and Eosin Y respectively with an attendant Azure B to Eosin Y ratio of 0.8:1.0 A similarly performing stain was a Giemsa sample from the ANDi Laboratory, Lagos, which gave a ratio of 1.1. These findings are at variance with previous studies that demonstrated satisfactory staining with Azure B to Eosin Y ratio ranging from 5:1 -10:1 (Wittekind, 1979). We, however, noted in this current study, a wide range of Azure B to Eosin Y ratio among the stain samples obtained from various laboratories and suppliers.…”

“…Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a;Friedrich K, Seiffert W, 1990;R. W. Horobin, 2011;WITTEKIND, 2008;Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and strength and direction of the association between each predictor and the likelihood of "satisfactory" staining outcome.…”

“…Ultraviolet spectroscopic and HPLC assays have revealed a high degree of variability of dye composition and the presence of RGE-irrelevant and RGE-antagonizing dyes in the stain samples analyzed in this study. Whereas the two most prominent dyes in our reference stain (Marshall et al, 1975c) were Azure B and Eosin Y, which are actually the only two dye species needed to produce the Romanowsky-Giemsa Effect (Friedrich et al, 1990a; Friedrich K, Seiffert W, 1990; R. W. Horobin, 2011; WITTEKIND, 2008; Wittekind DH and Gehring T, 1985), majority of the stain samples from various laboratories exhibited a wide range of variability with regards to the presence and proportion of these two essential dyes. This finding is similar to the results of previous commercial stains evaluation (Gordon B. Proctor and Richard W. Horobin, 1985; Horobin and Walter, 1987a; Marshall et al, 1975a; Schulte, 1991).…”

Quantitative diagnostic algorithm using endocytoscopy for superficial nonampullary duodenal epithelial tumors

“…Unexpectedly, the correlation between the median number of DSMs per 1 mm 2 in ECS images and the number of nuclei in histology was weak. As described in a previous report, we think the reason for this is that DSMs possibly include not only nuclei but also cytoplasm 20 . Furthermore, a recent retrospective study showed that ECS diagnosis with methylene blue provided a high accuracy rate, and double staining using crystal violet was not recommended for the diagnosis of SNADETs 10 .…”