In Veterinary Medicine

Current Issue
Previous Issues
Reprint Information
Back to The International Journal of Applied Research in Veterinary Medicine



Search Query
Click here for information on how to order reprints of this article.

Reduced Availability of Trimethoprim/Sodium Sulfadiazin and Doxycyclin Hyclate in Drinking Water Systems

Chris Vervaet, PhD

Jody Voorspoels, PhD

Filip Kiekens, PhD

Jean Paul Remon, PhD


Laboratory of Pharmaceutical Technology, Ghent University, Gent, Belgium.


KEY WORDS: Medication administration, dissolution, drinking systems, trimethoprim, doxycyclin hyctate



The availability of drugs in the reservoir and at the nipples of a drinking water system was documented following the addition of Trimazin®, a powder formulation containing sodium sulfadiazin (Na-SD) and trimethoprim (TMP) or doxycyclin hyclate powder (DoxHCl), to the system using a standard mixing methodology. The theoretical concentration of Na-SD, a highly water-soluble drug, was obtained immediately after addition of the Trimazin® powder, whereas the TMP concentrations were highly variable and did not yield their theoretical concentration due to the poor dissolution kinetics of TMP at the test conditions. The availability of DoxHCl was far below its theoretical concentration (200 mg/L), because the drug concentration in the water system did not exceed 30 mg/L. Adding 0.1% (w/v) citric acid to the water increased the DoxHCl concentrations to a maximum of 125 mg/L. This study shows that drug availability from a drinking water system can be well below the anticipated drug concentrations, due to solubility or dissolution rate problems. Therefore, the mixing methodology should be validated if drugs are administered via drinking water.


From an economic point of view, drug administration to a group of domesticated animals through medicated feed or drinking water is preferred over time-consuming and costly individual drug administration. Although the feed route is usually restricted to prophylactic treatment, medicated drinking water is often administered to sick animals for therapeutic purposes because they often lack appetite but continue to drink.1,2 However, water as a drug medium offers several challenges and limitations. The main obstacles are the drug’s solubility and stability in water. The solubility issue of powder formulations intended for use in drinking water systems is often solved by commercial formulations containing highly soluble salts or by the addition of suitable excipients. However, rapid and complete dissolution of the drug when adding the formulation to the water reservoir should not be taken for granted, because the water quality and design of the drinking water system (eg, characteristics of the mixing unit) could affect the drug’s dissolution kinetics. Because incomplete dissolution of the drug could result in a reduced availability, a suboptimal dosing regimen, and blocking the water system at the drinking points, drug addition to the drinking system should be validated by monitoring the drug concentration (not only in the water reservoir, but also at the drinking nipples). The goal of this study was to document the availability of drugs from drinking water systems on a pig farm, using daily used methodology to transfer and mix the drug formulation into the water reservoir. A commercially available trimethoprim-sulfonamide formulation was added to the water and the drug distribution throughout the water system was monitored; a second part of the study deals with the availability of doxycyclin hyclate at the drinking points.


Trimethoprim-Sodium Sulfadiazin

Materials: Trimazin® 30%, a water-dispersible powder for oral administration containing 25% (w/w) sulfadiazin (as a sodium salt), 5% (w/w) trimethoprim, and a-lactose monohydrate (as a filler), was provided by Kela Laboratoria (Hoogstraten, Belgium). Sodium sulfadiazin (Na-SD) and trimethoprim (TMP) used for analytical purposes were purchased from Sigma Aldrich (Bornem, Belgium). All reagents used during chromatographic analysis were analytical or HPLC grade.

Field experiment: One container (±1 kg) of the Trimazin® formulation (Kela) was dispersed in a bucket of lukewarm (±35˚C) water (±10L). The resulting suspension was immediately added to a stainless steel tank (maximum volume, 740 L) partially filled with well water (temperature, 8˚C). Both the bucket and the Trimazin®-container were rinsed with water from the tank. The drug/water mixture was homogenized for several minutes using the mixing unit built into the tank. The water reservoir was connected to a series of nipples (freely accessible for the pigs) through a set of pressurized tubes. If the pressure inside the tubes dropped below a threshold value due to water consumption by the pigs, a pump restored the pressure to a pre-set level. Simultaneously, the stirrer inside the tank was started to homogenize the drug mixture inside the tank.

Samples (10 mL) were taken from the tank just below the water surface and at 3 drinking points 0, 30 minutes and 1, 2, 3, 4, 7, and 9 hours after addition (and homogenization) of the drug mixture to the water system. At the 3 nipples, 2 fractions were collected at each sampling point. The first sample was taken immediately after opening the valve of the nipple, and the tubing was purged for 1 minute before the second sample was taken. Half of each sample volume was immediately transferred into borosilicate glass tubes (Corning Glass Works, Corning, NY, USA), the remaining part was filtered through a 0.22 µm cellulose acetate (CA) filter (Minisart, Sartorius, Vilvoorde Belgium). Awaiting HPLC analysis of the samples, all solutions were frozen immediately by transferring the sample tubes into an ethanol bath cooled to -70˚C with solid CO2. The Na-SD and TMP concentration in the samples was determined using a validated HPLC method.3

Laboratory experiments: The field experiment was simulated at lab-scale by dispersing 1g Trimazin® in 800 mL well water. These dispersions were put in a refrigerator (8˚C) and continuously mixed. Samples from these media were taken at the same time intervals as during the field experiment. The pH of each solution was measured after filtration through a 0.22 µm CA-filter (Minisart, Sartorius, Vilvoorde, Belgium).

Drug loss during filtration due to adsorption onto the filter medium was assessed by HPLC analysis of filtered and unfiltered samples (n = 3) of a Trimazin® solution (addition of 1 g powder to 800 mL well water, followed by 20 minutes of sonication to ensure complete dissolution of the samples).

Doxycyclin Hyclate

Materials: Doxycyclin hyclate and citric acid monohydrate were purchased from Ludeco (Brussels, Belgium). All reagents used during chromatographic analysis were analytical or HPLC grade.

Field experiment: During the study, 120 g doxycyclin hyclate (DoxHCl) was added to a stainless steel tank filled with 600 L tap water, yielding a theoretical DoxHCl concentration of 200 mg/L. In a second field experiment, 600 g citric acid was dissolved in the tap water available in the tank before the addition of 120 g DoxHCl. The water system was equipped with a pump for continuous circulation of the water through a closed loop system, providing the required mechanical agitation for homogeneous distribution of the drug in the water reservoir. The aqueous drug solution is transported toward the drinking points (the nipples situated in the boxes of the piglets) using a series of pressurized tubes. If the pressure inside the tubing decreased due to excessive water uptake, solution from the loop was pumped (using an automatic valve) into the tubes to maintain the required pressure at the drinking points. In the tank and at 3 nipples, samples (10 mL) were taken at specific time intervals after drug addition to the water system (0, 0.5, 1, 1.5, 2, 3, 4, 6, 8, 12, and 24 hours). These samples were immediately frozen at -20˚C until HPLC analysis. The doxycyclin concentration in the samples was determined using the validated HPLC method described by Santos et al.4


Trimethoprim and Sodium Sulfadiazin

As the exact volume of water in the drinking system, the exact fill weight of the Trimazin® formulation in the container, and the exact drug dosage in the formulation are unknown, the theoretical concentration of Na-SD was taken as the average drug concentration (n = 8) (406.2 ± 5.4 µg/mL) of the samples taken from the tank over the entire test period. Laboratory experiments showed that Na-SD (as a highly water soluble drug) dissolved instantaneously after its addition to water with limited mechanical agitation. The theoretical concentration of TMP (74.7 µg/mL) was calculated from the theoretical Na-SD concentration, taking into account the ratio of both drugs as mentioned on the label.

Figures 1A and B show the average concentration at the drinking points (n = 3) of sodium sulfadiazin (Na-SD) and trimethoprim (TMP), respectively, in relation to their theoretical concentration. The theoretical concentration (Ctheor) of Na-SD was obtained almost immediately at the nipples and was maintained throughout the entire test period, indicating an optimal availability of the drug. A similar profile was seen for the samples taken from the reservoir and at the nipples before purging. Although the purged samples yielded their Ctheor immediately after drug addition (t = 0 minutes), the Na-SD concentrations of the samples taken at t = 0 minutes and t = 30 minutes without purging were considerably lower, indicating that purging is required before an undiluted drug solution reaches the drinking points.

Unlike Na-SD, the concentration of TMP (Fig. 1B) at most sampling points was below its theoretical value and highly variable. The concentration profiles measured within the tank and at the nipples without purging showed the same trend (data not shown). Filtering these samples before analysis showed that the TMP-concentration decreased and reduced the variability, indicating that part of the drug remained in suspension throughout the entire experiment. The reduction in TMP recovery after filtration can partly be explained by the absorption of dissolved TMP molecules to the cellulose acetate-membrane during filtration of the samples (the loss ranged from 3.6% to 6.4%; no loss was detected for Na-SD), but it is mainly because of retention of the drug fraction in suspension throughout the entire experiment.

Simulation of the experimental conditions at laboratory scale using the same well water at 8˚C showed that Na-SD, as expected, yielded its Ctheor immediately after addition to the water. However Ctheor of TMP was only reached after about 2 hours of continuous mixing (Fig. 2; filtered samples). The higher TMP concentration of the unfiltered samples (Fig. 2) during the initial 2 hours of the test confirms that part of the drug remains suspended and that the dissolution kinetics of this drug were poor under these conditions (the temperature and pH of the well water were 8˚C and 8.62, respectively).5

The slightly lower TMP concentration of the filtered samples after 3, 4, and 7 hours are probably due to the limited absorption of TMP to the filter membrane. Because the close correlation between Ctheor and the actual concentration at the end of the test indicated that no chemical degradation occurred, these data confirm that the low TMP concentrations seen during the field experiment are due to the low efficiency of the intermittent mixing system used in this study. Even after 9 hours, about 25% of the drug was not dissolved; the remaining part was suspended throughout the system or sedimented in the reservoir and tubes. The data obtained under these specific test conditions stress the importance of an adequate mixing procedure (optimizing the duration and intensity of the mixing) to ensure maximum drug availability from the drinking system.

Doxycyclin Hyclate

Figure 3 shows that the actual concentration of DoxHCl was far below the theoretical concentration (200 mg/L). The maximum concentration in the tank and at the nipples did not exceed 30 mg/L. Instability of DoxHCl in water can not account for the loss in drug concentration shown by Santos et al.4 After 24 hours, about 85% DoxHCl was recovered from medicated drinking water stored in metal containers. Although DoxHCl is described in literature as a good water-soluble drug (1 part drug in 3 parts water; Merck Index), the limited availability of this compound is probably related to its solubility because laboratory experiments showed that the solubility of DoxHCl in the tap water was about 400 mg/L (at room temperature). Dissolving 0.1% (w/v) citric acid into the water before drug addition yielded higher DoxHCl concentrations. Its solubility in tap water increased to about 4 g/L. However, the concentrations obtained in the water system were still far below 200 mg/L (Fig. 3). Possible explanations for this observation are 1) that despite continuous homogenization of the water through a closed loop system, the mixing kinetics for such a large volume are insufficient; or 2) that if the drug dissolved completely, a large portion of the drug adhered to the inner surface of the water system (reservoir or tubing). Again, these data show the importance of monitoring the drug concentrations at various points in the water system and of designing a proper formulation to ensure complete drug dissolution and optimal availability of the drug.

Both field experiments indicated that drug availability from a drinking water system can be well below the expected drug concentrations, mainly due to solubility and dissolution rate problems. The mixing methodology used in combination with the specific water system should be validated to ensure complete drug dissolution, especially if poorly soluble drugs are administered. These data also stress the need for a proper cleaning procedure after administration of medicated drinking water because any residual drug (suspended in the system or adhered to the surfaces) will be released over a longer period, exposing the animals to a subtherapeutic dose. Because many drugs administered through drinking water systems are antibacterial drugs, resistance of the microorganisms to the drugs is a major issue under these circumstances. In addition, residuals in the water system might interact with subsequent drug therapies, possibly causing intoxication, as recently described by Croubels et al.6


1. Larrabee WL: Formulation of drugs for administration via feed or drinking water, in Formulation of Veterinary Dosage Forms. New York: Marcel Dekker, 1983:175–203.

2. Brooks PH: Water: Forgotten nutrient and novel drug delivery system. Feed Compounder 7:24–27, 1994.

3. Bonazzi D, Andrisano V, Di Pietra AM, Cavrini V: Analysis of trimethoprim–sulfonamide drug combinations in dosage forms by UV spectroscopy and liquid chromatography. Il Farmaco 49: 381–386, 1994.

4. Santos MDF, Vermeersch H, Remon JP, et al: Validation of a high-performance liquid chromatographic method for the determination of doxycycline in turkey plasma. J Chromatog Biomed Appl 682:301–308, 1996.

5. Meshali MM, El-Sabbagh H, Ramadan I: Simultaneous solubility and dissolution rate of sulphamethoxyzole and trimethoprim in binary mixtures. Pharmazie 39:407–408, 1984.

6. Croubels, S., Vrielinck, J., Baert, K., et al: A special case of acute tiamulin-salinomycin intoxication in pigs due to residual tiamulin four months after medication: Vl. Diergen. Tijdschr 70:54–58, 2001.



Figure 1. Concentration of sodium sulfadiazin (Na-SD) and trimethoprim (TMP) at the nipples (n = 3) of the drinking water system (containing well water at 8˚C), after the addition of ±1 kg Trimazin® formulation. The tubing was purged for 1 minute before collecting the samples. (A) Na-SD concentration of an unfiltered () sample. The theoretical Na-SD concentration (dashed line) was 406.2 mg/L. (B) TMP concentration of a filtered sample () and an unfiltered sample (). The theoretical TMP-concentration (dashed line) was 74.7 mg/L.


Figure 2. Dissolution profile of trimethoprim (TMP) in well water (8˚C, continuous stirring) as a function of time under laboratory conditions: mean TMP concentration (n = 3) of a filtered sample () and an unfiltered sample (). The theoretical TMP-concentration (dashed line) was 64.1 mg/L.



Figure 3. Mean concentration of doxycyclin hyclate (DoxHCl) at the nipples (n = 3) () and in the tank  (n) of the drinking water system (600 L) after the addition of 120 g DoxHCl at 0 minutes. The closed symbols represent the mean DoxHCl-concentrations (: at the nipples; n: in the tank) when 0.1% (g/v) citric acid was added to the water reservoir before drug addition. The theoretical DoxHCl-concentration (dashed line) was 200 mg/L.



©2000-2010. All Rights Reserved. Veterinary Solutions LLC
ISSN# 1559-470X