D-gluconic acid, compound with N,N''-bis(4-chlorophenyl)-3,12-diimino-2,4,11,13-tetraazatetradecanediamidine (2:1)

Endpoint summary

Description of key information

Considering the information available for chlorhexidine, i.e. its cross-sectional diameters and toxicokinetic studies indicating hindrance of uptake, its high water solubility, low experimental log Kow and predicted log Dow, and the results of experimental non-guideline BCF studies, it can be concluded that bioaccumulation is not likely. Estimated data on bioconcentration and bioaccumulation factors are consistent with experimental results.

In conclusion, this weight of evidence clearly demonstrate that Chlorhexidine has a low potential for bioconcentration and bioaccumulation in aquatic or terrestrial organisms. Further studies on bioconcentration and bioaccumulation of Chlorhexidine digluconate are scientifically not justified.

Additional information

Chlorhexidine has a low potential for bioaccumulation, based on a weight of evidence following additionally to BPR recommendations also the REACH guidances R.7 and R.11. This weight of evidence is presented hereinafter:

 

According to the BPR Annex II 9.1.4 first an estimationof the intrinsic potential for bioconcentration in aquatic organisms should be provided on the basis of physical and chemical properties (e.g. partition coefficient n-octanol/water). Secondly, if necessary, the bioconcentration should be determined in an experimental study. The experimental determination, however, may be omitted if a low potential for bioconcentration of the substance can be demonstrated on the basis of physico-chemical properties (e.g. log Kow < 3) or other evidence.Thereby all critical aspects of bioaccumulation such as ionic speciation, surface activity and metabolic transformation rates must be considered before experimental determination is considered unnecessary.

Substance properties with regards to ionic speciation, surface activity, and metabolic transformation rates

In aqueous solutions, Chlorhexidine digluconate will dissociate and the constituents will be present as non-ionic and ionic species, depending on the pH of the solution. As indicated by the pKa values for chlorhexidine and gluconic acid, chlorhexidine digluconate will be forming double protonated chlorhexidine cations and single deprotonated gluconic acid (gluconate) anions in aqueous solutions at biologically relevant pH values. Furthermore, the surface tension of rounded 64 mN/m at 20°C for a concentration of 1 g/L indicates negligible surface active properties. From the toxicokinetic studies on mammals (see as well below) it can be concluded that the small amounts of chlorhexidine which may be absorbed following oral administration are not metabolised to a measurable extent. For this reason, Chlorhexidine is not expected to be metabolized in significant amounts.

Screening criteria according to REACh guidance R.11

As mentioned above first an estimationof the intrinsic potential for bioconcentration in aquatic organismsshould be providedon the basis of physical and chemical properties (e.g. partition coefficient n-octanol/water). Phys-chem parameters can be derived by using different modelling approaches. Although the substances is ionic, appropriate estimations for the logDow (pH dependence of logPow; in the following logPow at pH7 is given) are available which are supported by experimental determinations as described in the following.

 

logD (ACD/percepta)

The ACD/percepta logD algorithm (www.acdlabs.com) predicts the distribution coefficient logD as the ratio of the sum of the concentrations of all forms of the compound (both neutral and ionized) in each of the two phases (octanol/water). Each of these forms has its own tendency to partition between water and n-octanol which is characterised by a pH independent partitioning coefficient. The logD prediction algorithm calculates such partitioning constants based on the fragmental algorithm of logP for the neutral form and a series of correction factors, taking into account aspects such as the type and position of ionization in the molecule. With this information it is possible to calculate logD as a function of the distribution of all molecular species. The latter is governed by pH as predicted by the pKa prediction algorithm.

For the validation of data estimated with the ACD software, a scientific report by Bassan et al. (2011) was prepared for EFSA showing that the logD could better describe the lipophilicity of the tested dataset than the logP.

Using the ACD/percepta program the log Dow for the chlorhexidine base has been calculated (Evonik 2020) using the pKa of the classic model to be 0.55, 1.58 and 0.0005 at pH 7 for the consensus, classic and GALAS model, respectively. Using the pKa of the GALAS model the logDow are even lower and has been calculated to be -1.44, -0.41 and -1.99. Overall, the logDow of the Chlorhexidine, which is at pH 7 a double protonated cation, is expected to be in the range of -1.99 until maximally 1.58.

 

This range is supported by experimental measurement of the logDow of the Chlorhexidine Base at pH 5 (acetate buffer) indicate a log Dow of 0.08 (Hansch et al 1995), as well as the experimental logKow of Chlorhexidine digluconate of -1.81 at 20.7 °C and pH 5.3- 6.6 (Degussa 2002).

In conclusion, all estimated and experimental derived logKow or logDow values at environmentally relevant pH values indicate that the partitioning coefficient is < 3, and thus an experimental study for bioaccumulation is scientifically not justified.

 

According to the REACh guidance R.11 and R.7 as well as the BPR guidance Vol IV parts B+C, the relative low logDow can be used with modelling approaches to estimate the bioconcentration (e.g. BCF_fishBCF_earthworm). Expected BCF_fish, biomagnification factor for fish-eating predators as well as theBCF_earthwormare very low.This is supported using a QSAR model integrated in the program ACD/LogD-Suite. The calculated BCF is 1.77 – 1.78 at environmental relevant pH (5 – 9) using the maximum log Dow of 1.58. Overall, the measured and estimated logDowsupports the experimental findings that Chlorhexidine does not significantly bioaccumulate in aquatic and terrestrial biota.

 

Screening criteria for air-breathing organisms according to REACh guidance R.11:

For air-breathing organisms, the screening criteria are Log Kow > 2 and log Koa > 5. Using again the logDow, the experimental and estimated values are all below < 2. This indicates that the substance will easily be excreted via urine if it will be taken up. The logKoa can be calculated by logKow – logKaw. The latter is the dimensionless Henry´s Law Constant. As the substance is double protonated at pH 7, and thus the Henry´s Law Constant is very low, the substance has a logKoa >> 5. This means that the substance cannot be exhaled if it will be taken up. However, oral and dermal uptake of Chlorhexidine is hindered (see below), and due to the low vapour pressure relevant concentration in air compartment and thus uptake via air is not expected.

Overall, the available information indicates that bioaccumulation in air-breathing organisms is not likely.

 

Non-lipophilic bioaccumulation according to REACh guidance R.11:

Based on the REACh guidance R.11 the non-lipophilic bioaccumulation has to be evaluated. Non-lipophilic bioaccumulation are known for fluorpolymers. However, Chlorhexidine does not contain any Fluor atom. The Canadian authority (Health Canada 2019) stated, that given the cationic nature of chlorhexidine, it is likely to interact with the negatively charged, phospholipid bilayer of cell membranes. Chlorhexidine potentially binds to protein (OECD QSAR Toolbox). However, based on the evaluation of the Canadian authority (Health Canada 2019), this is based on the ionic nature and chlorhexidine bridges between pairs of adjacent phospholipid head-groups. No empirical data on protein binding was found. Overall, the chlorhexidine moiety likely has a low potential to bioaccumulate.

 

Physicochemical and experimental indicators for hindrance of uptake according to REACh guidance R.11

According to the REACh guidance R.11, information on molecular size can be an indicator to strengthen the evidence for a limited bioaccumulation potential of a substance. Molecular size can be used in a Weight-of-Evidence approach together with other information. Very bulky molecules will less easily pass through the cell membranes. This results in a reduced BCF of the substance.

Calculations with the OECD QSAR Toolbox (OASIS) results for Chlorhexidine in a Deff and Dmax of 1.17 and 2.08 nm, respectively. This is supported by the calculation given by the Canadian authority (Health Canada 2019) using CPOP, the molecular weight of chlorhexidine (505.5 g/mol) and calculated cross-sectional diameters (D_effectiveand D_maximumof 1.1 nm and 2.1 nm, respectively; indicate that it is a relatively large molecule.

The Canadian assessment states additionally that investigations on fish bioconcentration factors (BCFs) and molecular size parameters show that the probability of passive diffusion via the gills decreases appreciably when the effective diameter of a chemical is greater than 1.1 nm and when the maximum diameter of a chemical is greater than about 1.5 nm (and much more so for molecules having a maximum diameter greater than 1.7 nm) (Dimitrov et al., 2002, 2003; Sakuratani et al. 2008) This is generally accepted and also implemented into the European guidance. According to REACh guidance R.11, it appeared that for compounds with a Dmax_averlarger than 1.7 nm the BCF value will be less than 2000.

Additionally, experimental indicators for hindrance of uptake should be considered in the weight of evidence. The mammalian toxicokinetic studies indicate that the uptake following oral or dermal application is limited. For the oral route an absorption rate of about 2 % of the applied dose can be assumed as most of the dose (> 90 %) is excreted unmetabolized via faeces; this is considered to represent unabsorbed material, while only very minor amounts (< 1.5 %) will be excreted via urine. The substance has a high affinity to the epithelia of the gastrointestinal tract. For the dermal route, all studies with humans are indicating a very low absorption with a maximum of about 0.02 % of the applied dose under different exposure conditions.

Overall, as the calculated value of Dmax of 2.1 nm using OASIS is significantly above the trigger value of 1.7 nm, and the mammalian toxicokinetic studies indicate a low absorption, the molecular size and experimental studies indicate the hindrance of uptake and thus supports that Chlorhexidine does not bioaccumulate.