Variability of Water Flow Rates and the Impact on Therapy Using Tetracycline Powder

by 5m Editor
21 July 2009, at 12:00am

The results presented at the 28th Centralia Swine Update by Glen Almond and colleagues at North Carolina State University indicate that medicating pigs with tetracycline in the drinking water is problematical and ineffective.


Considerable time, effort and expense are devoted to the treatment of pigs in commercial grow/finish barns. However, with large numbers of pigs in facilities, labour constraints, and the ability of some pathogens to spread to a high proportion of a herd, producers rely on therapeutic approaches for a ‘population’ of pigs. This approach typically involves the oral administration of antibiotics through the drinking water. Tetracyclines are widely used for water medication; however, the bioavailability of tetracyclines is low following oral dosing compared to intravenous injections of grow/finish pigs (Mevius et al., 1986; Nielsen & Gyrd-Hansen, 1996). Concerns regarding the administration of antibiotics through the drinking water prompted us to conduct a multiple phase study.

The objectives of the study were to:

  • evaluate drinker flow rates in commercial finishing barns
  • determine the relationship between water flow rate and plasma tetracycline concentrations in pigs housed in different pens in finishing barns
  • characterise plasma tetracycline concentrations in grow/finish pigs following the administration of tetracycline hydrochloride at various concentrations in drinking water, and
  • establish pharmacokinetic parameters for plasma tetracycline concentrations in grow/finish pigs consuming water medicated with tetracycline hydrochloride.

Materials and Methods

Drinker Flow Rates in Finishing Facilities: The first phase of the project was conducted on 13 finishing sites (farms). All finishing sites were within the same system of one company. The number of barns (n=100; 36-44 pens/barn) per site ranged from five to nine. Six of the sites used drinkers (Trojan®; 2/pen) mounted on the pen divider. The other seven sites used swing-type drinkers with two water nipples (Trojan®) per drinker line. Water flow rates (litre/min) were determined for all drinkers in the previously described pens and barns.

Water Medication with Tetracycline Hydrochloride: The second phase of the study was conducted on two sites from the original 13 sites. The first site had low and uniform flow rates (1.44 ± 0.65 l/min). In contrast, the second site had high and variable flow rates (2.63 ± 1.17 l/min). Other than the flow rates, the barns had similar numbers of pens, stocking density and age of pigs (11-12 weeks of age). For each barn, 10 pigs were randomly selected and ear tagged in the designated sick pen and in four additional, randomly selected pens (total 50 pigs/barn). Tetracycline hydrochloride soluble powder (AmTech -324®) was diluted (1:128) in a stock solution for the administration through a Dosatron® medicator. Farm personnel set the medicator to deliver tetracycline hydrochloride (22 mg/kg BW). The medicated water was provided to the pigs for three days. Water flow rates were determined daily for each drinker in the pens.

Blood samples were collected from pigs prior to medication (Time 0) and at 4, 8, 24, 48 and 72 hours after the initiation of water medication. Water samples were collected from the drinkers at the same time points. Water and plasma samples were analyzed with ultra-high performance liquid chromatography (UPLC) to quantify tetracycline concentrations.

Phase 3 Procedures, Animals and Treatments: A randomized design experiment was performed with 24 barrows (7-8 weeks of age). Pigs were weighed four days prior to the initiation of treatments and at the completion of the study. Pigs were randomly assigned to treatment (n=6 pigs/treatment) with 0, 125, 250, or 500 ppm tetracycline (324 g/lb tetracycline, AmTech Inc.) in the drinking water.

Pigs were housed in individual pens (1 m x 2.2 m) at the University Swine Educational Unit. Each pen was equipped with one drinker and one feeder. For each pen, a 20 litre carboy was used as the sole source of drinking water and water medicated with tetracycline hydrochloride. Aratowerk 80™ drinkers were used to minimize water spillage.

Sample Collection and Analysis: Blood samples were collected from each animal immediately prior to treatment and 4, 8, 12, 24, 32, 48, 56, 72, 80, 96 and 104 h after treatment. In addition, water samples were collected from each pen at the time of blood collection. All plasma and water samples were analyzed via HPLC techniques.

Conventional PK analysis was performed using WinNonLin (Pharsight, Mountain View California). Concentration of steady state (Css), time of steady state (Tss), maximum plasma concentration (Cmax), observed time of maximum plasma concentration (Tmax.), and area under the curve (AUC) were determined. In addition, the mean residence time (MRT) was based on concentrations at 32-48 hours (the elimination curve at steady state) and the half-life was defined as the time (h) to clear one-half of the concentration of the drug from the body. Bioavailability (F) also was determined.

Results and Discussion

Drinker Flow Rates: The initial linear regression of flow rates indicated that farm, barn, drinker type and medication were significant (P<0.01) in the overall model. Conversely, pen, sick pen and drinker placement were not significant. Drinker flow rates varied among farms and barns (Table 1). The mean drinker flow rates ranged from 1.44 to 2.77 litres/min. Variations in flow rates were evident in barns, and the pen-to-pen variation within a barn was quite dramatic. Differences among pens were not apparent (Figure 1). Evidently, differences in flow rate among pens within barn reflected the drinkers and not the pressure in the water lines. Water medication (typically tetracycline) was being used in at least one barn per farm. Flow rates were less (P<0.05) from drinkers with medicated water (1.96 ± 0.03 litres/min) than from drinkers without medication (2.3 ± 0.01 litres/min). Flow rates also differed (P<0.05) between drinker type. The swing drinkers provided 2.17 ± 0.015 litres/minute (n=4,584 drinkers). In contrast, the flow rate for mounted drinkers (n=2,538 drinkers) was 2.43 ± 0.02 litres/minute.

Table 1. Drinker flow rates (litres/min) for 13 finishing farms (5-9 barns/farm). Drinker types were either mounted (M) on the pen divider (1-2 drinkers/pen) or swing-type drinkers (S; 2 nipples/drinker). The numbers of pens ranged from 36-44 per barn. Within column, values with different superscripts differ (P < .05).
Farm Drinker Type No. Barns No. Drinkers Flow Rate (Mean + SEM)
1 M 8 672 2.33 ± 0.038 de
2 S 5 440 2.37 ± 0.055 ab
3 S 9 648 1.44 ± 0.025 g
4 S 9 648 1.85 ± 0.028 g
5 S 9 648 2.63 ± 0.046 bc
6 M 6 424 2.32 ± 0.073 e
7 S 9 704 2.48 ± 0.034 cd
8 M 8 272 2.58 ± 0.040 abc
9 S 9 648 2.06 ± 0.037 f
10 M 8 702 2.30 ± 0.030 e
11 S 7 560 2.20 ± 0.042 ef
12 M 7 504 2.77 ± 0.050 a
13 M 7 252 2.62 ± 0.062 abc

Figure 1. Drinker flow rates (mean + SEM) by pen for 13 farms (n = 100 barns). Only pens 37 and 21 differed.

Phase 2 - Plasma Tetracycline Concentrations: Plasma tetracycline concentrations increased from time 0 to reach peak concentrations at 8 h and 48 h after the initiation of water medication (Figure 2). The time of the peak was evident at 8 h in Farm 38 for all pens. Tetracycline concentrations did not differ among sick pens and healthy pens. In both farms, two pigs died in the sick pens before the completion of the study. Tetracycline was not detected in the plasma of these particular pigs before death, thereby indicating that these pigs likely failed to consume water.

Figure 2. Plasma tetracycline concentrations (mean + SEM) in pigs (12 weeks of age; 10/pen; five pens/barn) treated with tetracycline hydrochloride as a water medication. The top figure shows the overall means (+SEM) for the time points for all pigs and all pens. The bottom figure shows the means (+SEM) for each time by barn. Points with different superscripts differ (P<0.05).

Overall, the mean plasma tetracycline concentrations, excluding concentrations at time 0, were greater (P<0.05) in Barn 39 (0.27 ± 0.01 μg/ml) than in Barn 38 (0.16 ± .006 μg/ml). It should be noted that considerable variation was evident in the tetracycline concentrations among individual pigs. Drinker flow rates were greater (P<0.01) in Farm 39 (3.9 ± 0.04 litres/min) than in Barn 38 (1.4 ± 0.03 litres/min). Subsequent Spearman rank correlations indicated that drinker flow rate (r=0.2) and farm (r=0.23) were correlated (P<0.01) to tetracycline concentrations, while pen was not significant.

The initial correlations to drinker flow rate are misleading. The correlations indicated that additional factors influenced plasma tetracycline concentrations. Thus, it was not surprising that tetracycline concentrations in the drinker water were 85.03 + 2.1 and 153.1 ± 7.4 μg/ml for Barns 38 and 39, respectively. Therefore, the pigs in Barn 39 had access to greater drinker flow rates and higher water concentrations of tetracycline than the pigs in Barn 38. For both barns, it is likely that mixing errors or malfunctioning medicators contributed to the modest tetracycline concentrations in the water.

Based on previous reports (Mroz et al., 1995, Almond 2002), the pigs in the present study likely consumed between two and three litres of water each day. With water tetracycline concentrations of 153 μg/ml, the pigs in Barn 39 would have consumed 306-450 mg tetracycline/pig/day. In contrast, pigs in Barn 38 would receive less than 255 mg. Although these doses of tetracycline may be suboptimal, the provision of tetracycline chloride at 400 mg/litre and 800 mg/litre in drinking water resulted in serum concentrations of less than 0.6 μg/ml (by 60 hr) and 0.8 μg/ml (10 hr), respectively (Luthman et al., 1989). Furthermore, a single oral dose of 40 mg/kg in 200 ml water resulted in serum concentrations of less than 1 μg/ml in non-fasted animals.

Collectively, the present study confirmed previous observations (Luthman et al., 1989, Nielson and Gyrd-Hansen, 1996) that medication of drinking water provides inconsistent and low plasma concentrations of tetracycline. In fact, the inclusion of 0.55 g of OTC/kg feed to 30 kg BW pigs resulted in similar plasma OTC concentrations (Hall et al., 1989) as noted in studies using water medication.

The relatively low plasma concentrations achieved by water medication with tetracycline raises concerns about absorption of the antibiotic. It is obvious that feed interferes with absorption since fasted animals had greater plasma tetracycline concentrations than non-fasted animals (Luthman et al., 1989). Secondly, treatment of pigs with doxycycline (DOX; 90 µg/ml of drinking water) resulted in mean plasma concentrations from 0.83 to 0.96 µg/ml (Croubels et al., 1998) to 1.37 ± 1.2 µg/ml (Prats et al., 2005). Despite marked inter-animal differences, these plasma concentrations were higher than the minimum inhibitory concentrations (MICs) described for bacterial pathogens of the respiratory tract (Prats et al., 2005). Evidently, the absorption of DOX is sufficient, at the aforementioned concentrations in controlled experiments, to provide therapeutic plasma concentrations.

Phase 3 – Plasma Tetracycline Concentrations In Individual Pigs: As shown in Table 2, initial body weights, total body weight gain, drinker flow rate, and daily water use did not differ among groups. Total protein and PCV of the blood was not affected by treatment. The HPLC analysis indicated that the water concentrations of tetracycline were at the desired levels of 125, 250 and 500 ppm.

Table 2. Descriptive statistics of parameters assessed during the study.
Means did not differ among treatment groups. Pigs were weighed four days prior to the initiation of treatments and again after the last blood samples were collected. Thus, total gain is for a 9 day interval. One animal in the 125 ppm group used 12 litres of water each day and was excluded from the analysis.
Parameter 0 Treatment Group SEM
125 ppm 250 ppm 500 ppm
Starting wt (lb) 43.6 41 42.9 42.1 1.5
Finishing wt (lb) 57.3 54.8 55.8 55.2 2.1
Total gain (lb) 13.7 13.8 13 13 1.2
Drinker flow rate (ml/min) 635 628 602 653 37.7
Daily water use (ml/pig) 3921 2818 2828 2851 647
Blood - total protein (g/dl) 5.6 5.7 5.7 5.8 0.26
PCV (%) 34.5 39.7 37 38.3 2.0

Animals in the 500 ppm group had the highest plasma concentrations, with steady state concentrations at 0.74 ppm, while the pigs receiving 250 ppm and 125 ppm had tiered concentrations below this (Table 3). Similarly, the area under the curve (AUC) total was higher in the 500 ppm group than in the 125 and 250 ppm groups. The Tmax was earlier in the pigs treated with 500 ppm than the other groups. Due to the variation in tetracycline concentrations, few means differed among the three tetracycline groups. In general, the mean concentrations were greater (P<0.05) in the 500 ppm group than in the 125 ppm group (Figure 3). The notable exceptions were at 24 and 48 h. There were no differences in the mean plasma concentrations between the 125 ppm and 250 ppm groups. However, it was noted that a doubling in the tetracycline dose did not double the expected average plasma concentration seen in a treatment group. Therefore, as tetracycline dose is increased, the steady state plasma concentrations increased linearly, but to not as rapidly as dose increased.

Table 3. Pharmacokinetic parameters (means) for the three treatment groups
Parameter Treatment Group
125 250 500
Css 0.33 0.47 0.74 ug/ml
Tss 32 32 32 hours
Cmax 0.80 1.26 1.29 ug/ml
Tmax 80 56 12 hours
AUCtotal 30.71 44.93 73.74 ug*h/ml
AUC32-48 3.54 6.55 10.46 ug*h/ml
MRT 6.64 6.61 6.44 hours
Half-life 4.60 4.58 4.46 hours
Clbody 0.14 0.12 0.18 L/kg/h
Vdss 0.919 0.803 1.188 L/kg
F 5.2 3.3 3.4 %
All parameters were calculated in WinNonLin (Pharsight).
Css, concentration of steady state; Tss, time of steady state.
Cmax, the maximum plasma concentration; Tmax, the observed time of peak plasma concentration. Cmax and Tmax were calculated from individual animals.
AUC, area under the curve. The AUC total is not extrapolated to infinity.
MRT (mean residence time) was based on 32-48 hours (the elimination curve at steady state).
Half-life, the time (h) to clear one-half of the concentration of the drug from the body.
Cl body, clearance rate from the body.
Vdss, volume distribution of steady state.
F, bioavailability.

Figure 3. Mean (+STD) plasma tetracycline concentrations in the control and three treatment groups

Twice daily clinical evaluations did not reveal clinical signs of disease in the pigs prior to or during the study period. Thus, the pharmacokinetic evaluation of the three levels of tetracycline in water was conducted on normal animals without the need to account for the influence of disease on water uptake. Since the blood TP and PCV values were within normal limits, it can be assumed that the frequent blood sample collection did not interfere with the hemodynamics of the pigs. Based on the failure to detect differences in weight gains, it was evident that five days of water medication did not influence growth. As anticipated, the Aratowerk™ drinkers provided uniform flow rates of water to all pigs. In contrast, daily water use was reduced when the water was medicated with tetracycline. Regardless of concentration, it was evident that medication of water with tetracycline impairs water use and presumably consumption. This must be taken into consideration if the tetracycline is intended to be used for therapy.

When pigs received tetracycline water medication, plasma concentrations were less than 1 ppm, even under non-competitive circumstances. The Vdss values were somewhat lower than values previously reported for tetracycline (Nielsen and Gyrd-Hansen, 1996) and oxytetracycline (Mevius et al., 1986); however, previous studies used 20 to 45 mg/kg, administered as a bolus dose through a stomach tube or as a drench. Although the Vdss are low in the present study, there is reasonable distribution in the body. In regard to other pharmacokinetic parameters, the Tmax values indicated that greater than 48 hours is required to reach the Cmax values with 125 and 250 ppm tetracycline. If the Cmax values were sufficiently elevated to be therapeutic, the pigs must be medicated for a minimum of two days. The failure to achieve plasma tetracycline concentrations greater than 1.0 ug/ml is consistent with previous studies that used bolus oral doses of 45 mg/kg (Nielsen and Gyrd-Hansen, 1996). Regardless of the method of administration, it is evident that water medication provided sub-therapeutic plasma concentrations, as observed in phase 2, which was conducted in commercial finishing barns. Furthermore, the bioavailabilities for the three doses, albeit similar to the aforementioned studies, are indicative of reduced absorption. The causes of the reduced absorption are speculative; however, bioavailability was higher in fasted pigs than in fed pigs (Nielsen and Gyrd-Hansen, 1996). Evidently, the presence of feed in the gastrointestinal tract impairs tetracycline absorption.

The therapeutically active plasma concentrations for tetracycline or other antimicrobial agents are speculative since the susceptibility of various bacterial pathogens are highly variable. Based on the minimum inhibitory concentration (MIC) for Escherichia coli AT2259 (quality control strain), an MIC of at least 1-4 ppm (1-4 μg/ml) is necessary to prevent growth of bacteria. The tetracycline MIC’s (50 per cent and 90 per cent) for other bacterial pathogens, such as Actinobacillus pleuropneumoniae and Streptococcus suis, typically range from 8 to more than 32 μg/ml (Salmon et al, 1995).

Based on this information, it is apparent that water medication with tetracycline hydrochloride does not offer a therapeutic level of drug in the plasma to treat a septicaemia. Tissue levels (lung, liver, kidney) of this medication will have to determined in the future. Despite the provision of optimal housing conditions, elimination of pig-to-pig competition and doubling the manufacturer’s recommended dose of tetracycline hydrochloride, the plasma tetracycline concentrations apparently were less than therapeutic levels.

Finally, the present results confirm our concerns, raised in the previous investigation, that medicating pigs with tetracycline in the drinking water has inherent problems and questionable efficacy.

Literature Cited

Almond, G.W. 2002. Water: Optimizing performance while reducing waste. Proc. NC Pork Expo.
Mevius, D.J., Vellenga, L., Breuking, H.J. et al. 1986. Pharmacokinetics and renal clearance of oxytetracycline in piglets following intravenous and oral administration. Vet. Quarterly 8:275-284.
Nielsen, P. and Gyrd-Hansen N. 1996. Bioavailability of oxytetracycline, tetracycline and chlortetracycline after oral administration to fed and fasted pigs. J. Vet. Pharmacol. Therap. 19:305- 311.
Plumb, D.C. 1999. Veterinary Drug Handbook. 3rd Edition. Iowa State University Press: Ames, Iowa.
Carr, J. 2002. Water systems – troubleshooting common mistakes. ISU Swine Disease Conference for Swine Practitioners. 2002.
Pijpers, A., Vernooy J.A.C.M., van Leengoed L.A.M.G. and, Verheijden J.H.M. 1990. Feed and water consumption in pigs following an Actinobacillus pleuropneumonia challenge. Proceedings of 11th IPVS Congress, Lausanne, Switzerland, 6:39.
Luthman, J., Jacobson, S.O., Bengtsson, B. and Korpe C. 1989. Studies on the biovailability of tetracycline chloride after oral administration to calves and pigs. J. Vet. Med. A. 36:261-268.
Hall, W.F., Kniffen T.S., Bane, D.P. et al. 1989. Plasma concentrations of oxytetracycline in swine after administration of the drug intramuscularly and orally in feed. J. Am. Vet. Med. Assoc. 194:1265-1268.
Croubels, S., Baert, K., De Busser, J. and De Backer, P. 1998. Residue study of doxycycline and 4- epidoxycycline pigs mediated via drinking water. Analyst 123:2733-2736.
Prats, C., El Korchi G., et al. 2005. PK and PK/PD of doxycycline in drinking water after therapeutic use in pigs. J. Vet. Pharmacol. Therap. 28:525-530.
Salmon, S.A., Watts, J.L., et al. 1995. Comparison of MICs of ceftiofur and other antimicrobial agents against bacterial pathogens of swine from the United States, Canada, and Denmark. J. con Micro. 33:2435-2444.
Mroz, Z., Jongbloed, A.W., Van Diepen, J.T.M. et al. 1995. Excretory and physiological consequences of reducing water supply to nonpregnant sows. J. Anim. Sci. (Suppl 1). 73:213.

Further Reading

- You can view other papers presented at the Centralia Swine Research Update 2009 by clicking here.

July 2009