Sage Journals: Discover world-class research

Abstract

The current study investigated how the gastric evacuation rate (GER) was affected after surgically introducing dummies of a blood flow biotelemetry system into the abdominal cavity of Atlantic cod, Gadus morhua. Gastric evacuation experiments were performed two and 10 days postsurgery on surgically implanted and control G. morhua force-fed sandeel, Ammodytes tobianus. The results were compared with previously obtained estimates from unstressed conspecifics voluntarily feeding on a similar diet. After two days, GER was significantly lower in the group of fish with the dummy implants compared with the control group, but following 10 days of recovery no significant difference was seen between the two groups. The difference between implanted and control fish observed two days postsurgery may have resulted either from surgery, postsurgical stress and/or the presence of the implant. The conclusion is that 10 days of postsurgical recovery will stabilize GER in G. morhua, thus indicating that at this point the implant per se did not affect GER. Both the fish with surgical implants and controls in this study evacuated their stomachs much slower and with much higher interindividual variation compared with G. morhua feeding voluntarily on similar prey items. The lower GER and higher interindividual variation for force-fed fish indicate that handling, anaesthetization and force-feeding impair GER and that individual fish respond differently to the suppressing effects.

Keywords

Postsurgical recovery gastric evacuation Gadus morhua implant biotelemetry

Invasive procedures are often necessary for the physiologist to understand the mechanisms behind various physiological parameters. In fish, measurements of cardiovascular parameters such as blood flow and blood pressure involve handling, anaesthesia and surgical implantation of catheters and flow probes in and around vessels and sutures to close the incisions. In traditional laboratory experiments, standard bench top blood pressure and blood flow measuring units are used, which means that the animals are ‘hardwired’ to the equipment during the experiments. Catheters and leads from flow probes penetrate the body wall and increase the risk of infection. Furthermore, to prevent tangling and ripping of wires the animals are most often confined in small containers, which severely limit the range over which they can move with the risk of introducing confinement stress. The animals are usually left for 18–48 h before experimentation is initiated, the main reason for such shorts periods being the high risk of infection after an extensive surgery.¹ In cases where catheters are used, limits on the time where these will remain open may also be decisive. However, unless appropriate recovery time is ensured following surgical procedures and instrumentation, major concerns are the validity of data and ethical considerations. In a study on rainbow trout Oncorhynchus mykiss using bioelectric potential recordings to measure heart rate, fH (i.e. a non-invasive method) the fish were handled and then left to recover for at least three days after which their basal heart rates were much lower than those presented in previous studies using traditional methods involving surgical procedures.² A lower heart rate is usually indicative of lower stress levels pointing to the fact that these animals were not as stressed due to confinement and the use of ‘hardwired’ techniques. Obviously, any misinterpretation of data may have important consequences for our understanding of the physiology of fishes and researchers should be encouraged to seek refinements of techniques to ensure more valid results as well as better animal welfare.

The rapid development of telemetric devices opens new research areas that have not been possible with hardwired animals and during the last few decades there have been great advancements made in applications of biotelemetry in fish physiology. Biotelemetry allows data collection from specimens that move around freely and can behave in a more natural way. In the telemetrically instrumented animal, the risk of infections is lower since there are no skin-penetrating wires. Furthermore, long-term recovery and recording periods are possible in reducing stress from handling, making this an attractive alternative. Altogether, this provides a number of welfare advantages and the opportunity to obtain a higher quality of the measured variables and more reliable routine values. A recent study on white sturgeon (Acipenser transmontanus) showed that routine heart rate of freely moving biotelemetric instrumented fish was lower compared with hardwired, confined conspecifics as well as with previous values reported for white sturgeon.^3,4 The downside is that the animals are carrying the entire recording/transmitting unit, increasing their body mass and possibly impairing the capacity to generate thrust by altering the ability to bend the body. It may furthermore interfere with the internal organs if placed in the body cavity. Nevertheless, a fully implantable system maximizes the likelihood that the studied fish will be treated normally by surrounding fish⁵ and minimizes the risk of infection and expulsion.⁶ Even the presence of a small, protruding antenna has been shown to cause adverse tissue reaction at the antenna exit^7,8 and to elicit aggressive attacks from other individuals of Atlantic salmon smolts.⁵

We are looking to use a fully implantable biotelemetry system^4,9 to study blood flow distribution in Atlantic cod, Gadus morhua L. voluntarily feeding and moving between different habitats. Before employment of such a system it is, however, imperative to evaluate appropriate postsurgery recovery time and possible adverse effect of the telemetric implant. Using a dummy version of the dual blood flow biotelemetric system used by Gräns et al.,^4,9 the aim of the present study was to examine the effects on the gastric evacuation rate (GER) of handling, surgery and instrumentation of G. morhua with a dual channel blood flow telemetric system following short-term (2 days postsurgery) as well as prolonged (10 days postsurgery) recovery. Both the surgically implanted and control fish were force-fed to standardize the amount of food and the time of feeding. The ability to evacuate the force-fed meal was determined and compared with estimates obtained from unstressed conspecifics that fed voluntarily.

Materials and Methods

Fish

Wild G. morhua ranging in total body mass from 290 to 815 g were caught in late 2009 by fyke net in the vicinity of the Marine Biological Laboratory in the northern part of Øresund, Denmark. The fish were transferred to the laboratory where they were kept in 9–10°C re-circulated, aerated seawater. The tank was covered by tarpaulin to avoid unnecessary disturbance of the fish. Light/dark conditions were 16:8 h. All fish were acclimatized to laboratory conditions for a minimum of 12 weeks before experiments were initiated and fed ad libitum three times a week with chopped herring (Clupea harengus L.) or Raitt's sandeel (Ammodytes marinus Raitt). Before the experiment the fish were starved for five days to ensure that their stomachs were empty.

Preoperative care

Individual G. morhua were anaesthetized in 15 L of water containing 0.15 g/L 3-aminobenzoicacid ethyl ester (MS-222) until gill ventilation ceased. Each fish was weighed and its length measured before it was positioned with the ventral side facing up on an operating table covered with wet sponges. Anaesthesia was maintained during the surgery by pumping oxygenated sea water containing MS-222 (0.075 g/L) over the gills. All surgical instruments, the dummy implant and associated leads were sterilized using a cold steriliant (Cidex, Johnson & Johnson Company, New Brunswick, NJ, USA).

Dummy implant

Our long-term goal is to implement the fully implantable biotelemetry system used by Gräns et al.^4,9 This system weighs 35 g (total mass of implant and battery) in air and we aim for a transmitter:fish mass ratio of 3–4%. Consequently, the dummy implant was constructed so that its total mass equalled 19 g in air, thus, corresponding to an average of 3.7 (0.3)% of the mean body mass of the present experimental G morhua. It was made of silicon-coated stainless steel (L = 35 mm, H = 5 mm, W = 16 mm) with an attached silicon-coated dummy battery made of plastic tubes (L = 18 mm, D = 11 mm) and two silicon leads with a probe cuff each, and the implant fitted easily within the abdominal cavity of the fish.

Surgery

The splanchnic circulation in G. morhua is derived from the celiacomesenteric artery (CoMe). The CoMe is the largest single branch of the dorsal aorta and branches to form the coeliac artery and mesenteric artery. The coeliac artery supplies parts of the cardiac stomach, the pyloric stomach, the proximal intestine and parts of the distal intestine, while the mesenteric artery supplies the remainder of the gastrointestinal system, the gallbladder and the spleen.¹⁰ The surgical procedures were performed on 10 fish (body mass 550 [144] g and length 38.7 [3.2] cm) and done in accordance with the guidelines described in permission 2010/561–1812 from the Danish Ministry of Justice. For placement of the first dummy probe, a 2 cm long incision was made alongside the basibranchial bone left of the midline, and the liver carefully retracted, after which the CoMe artery was exposed and cleared. The dummy implant and battery were then carefully placed in the abdominal cavity and the retracted organs restored to their places of origin. Furthermore, an individual passive integrated transponder (PIT) tag (12 mm × 2 mm, ISO standard 11784/11785, FDX-B) was placed in the body cavity for later identification of the fish. To access the ventral aorta for placing the second dummy probe, a 0.5 cm ventral incision was made posterior to the gill juncture and carefully, without disrupting the pericardium or damaging any vessels, the dermal and subdermal musculature and connective tissue were separated using blunt dissection tools to expose the ventral aorta. The probe was tunnelled under the skin from the abdominal cavity to the incision above the ventral aorta (see Figure 1 for schematic overview of positioning of the system). The wires from the probes were anchored with a single stitch of 3/0 silk suture in the subdermal muscle tissue and the two incisions were closed using sterile monofilament prolene 3–0 suture. Subsequently, the cod was given a subcutaneous injection of 0.4 mg/kg [1 mmol/L] butorphanol (Torbugesic, Fort Dodge, IA, USA) for postoperative pain relief in addition to an injection with 10 mg/kg [1 mmol/L] enrofloxacin (Baytrilo, Bayer, Shawnee Mission, KS, USA) antibiotics, to minimize the risk of infection. Finally, to reduce incidence of oomycete infections, povidone iodine powder was applied to the closed incision before returning the fish to recovery water. The surgical procedure including anaesthesia and awakening took approximately 20 min for each fish and when ventilation and locomotion had been re-established, the fish were returned to their home tank. All but one fish survived the surgery.

Figure 1

Schematic drawing of a Gadus morhua showing the placement of the dummy implant and battery (both drawn to scale) in the abdominal cavity and the flow probes belonging to it. The celiacomesenteric artery branches off the dorsal aorta (DAo) and subsequently divides into the coeliac (ACo) and the mesenteric artery (AMe). VAo: ventral aorta; STO: stomach

The control fish (n = 8, body mass 518 [155] g and length 37.1 [2.7] cm) were anaesthetized as described above, and subsequently weighed, length measured and individually PIT tagged. The tags were placed in the abdominal cavity, using a syringe implanter, taking care not to harm any organs. The control fish were likewise left to recover before being returned to their home tank.

Food

Ammodytes marinus constitutes a natural and large part of the diet of wild G. morhua and one specimen per cod was used as the experimental meal. To minimize variation in size and condition, specimens of A. marinus from a single commercial batch (TripleNine, Hvide Sande, Denmark) were used for force-feeding. The energy density of these prey fish was determined by bomb calorimetry (IKA C-7000 bomb calorimeter [www.ika.net]) on a representative sample after drying it at 60°C until constant mass according to the procedure described in Pedersen & Hislop.¹¹

Gastric evacuation experiments

The first experiment was initiated 48 h postsurgery. On the day of experiment, A. marinus were quickly thawed in running water, and those of similar total lengths were selected. These were subsequently dabbed dry with paper towel and weighed, and individuals within a mass range of 9.4–15.7 g were chosen (corresponding to 2–3% of the body mass of the G morhua). This food ration was chosen from the largest daily feeding rates observed in wild G. morhua.¹² In random order individual cod were anaesthetized by placing them in 15 L of water containing 0.07 g/L MS-222. They were then force-fed by gently pushing a whole A. marinus, bent in the middle, through the oesophagus and into the stomach. Immediately thereafter, the fish were returned first to recovery water and subsequently to their home tank. The exact time of force-feeding was noted for each fish, as was the length and mass of the prey.

Stomach contents were recovered 22–26 h post-force-feeding by stomach flushing following anaesthetization (0.07 g/L MS-222). The stomach content from each fish was collected on a filter with a mesh size of 200 μm, and gently patted with a moist paper towel and weighed. The exact time for recovery of stomach content was noted for each cod.

The fish were subsequently left to recover for seven days in their home tank, after which the above described protocol was repeated in a second trial.

Calculations and statistical analyses

In accordance with the cylinder model of Andersen and Beyer,¹³ the GER of stomach contents in G. morhua (as well as in a variety of other piscivorous fishes) can be described independently of meal size by the current mass of stomach contents S_t (g) together with the rate parameter ρ:

\frac{d S_{t}}{d t} = - ρ \sqrt{S_{t}}

(1)

Integrated from time 0 of force-feeding to time t (h) of recovery of stomach contents, the solution to Equation (1) is G. morhua evacuating its stomach contents according to the relationship

\sqrt{S_{t}} = \sqrt{S_{0}} - \frac{1}{2} ρ t; 0 \leq t \leq t_{end} = 2 ρ^{- 1} \sqrt{S_{0}}

(2)

where t_end is the time of complete evacuation of the prey.

The ability to evacuate the stomach contents is then described by ρ, which can be calculated for each individual G. morhua by the reorganization of Equation (2):

ρ = 2 (\sqrt{S_{0}} - \sqrt{S_{t}}) t^{- 1}

(3)

The rate parameter ρ depends on predator (total) length L (cm), temperature T (°C) and prey energy density E (kJ/g). The effects of these variables were estimated by Andersen¹⁴ for different gadoids fed fish prey, and the following relationship was obtained:

ρ = ρ_{LTE} L^{1.44} e^{0.078 T} E^{- 0.86}

(4)

The basic rate parameter ρ_LTE specifies the general ability of the predator to digest and evacuate a specific prey type from the stomach (or reversed: it indicates the resistance of the specific prey to the digestive processes in the stomach). The value 0.00142 ± 0.00011 (estimate ± SD) of ρ_LTE was obtained from an experiment by Andersen¹⁴ on G. morhua that fed voluntarily on lesser sandeel Ammodytes tobianus L. The temperature (10.3°C) and mean body size of G. morhua (41 cm; 664 g) were similar to those of the present study. This value of ρ_LTE was used to decide the evacuation time t of c. 24 h employed in this study, which was shorter than t_end and yet long enough to ensure that a significant part of the meal was evacuated from the stomach prior to recovery of the remains.

G. morhua used in the present study varied to some extent in size (Table 1). ρ_LTE was therefore used to represent the ability of individual G. morhua to evacuate their force-fed meals of A, marinus. By using this parameter, the results could further be compared directly with the above estimate obtained from unstressed G. morhua that fed voluntarily. Combining Equations (3) and (4), the value of ρ_LTE for each individual G. morhua was calculated by

ρ_{LTE} = 2 (\sqrt{S_{0}} - \sqrt{S_{t}}) t^{- 1} L^{- 1.44} e^{- 0.078 T} E^{0.86}

(5)

Table 1

Basic data on gastric evacuation experiments (mean [SD]) on Gadus morhua

		Predator G. morhua			Prey Ammodytes marinus
Group and experiment	L (cm)	Body mass (g)	Time t (h) for recovery of force-fed meal	Original body mass (g)	Recovered body mass at time t	Energy E (kJ/g)	Observations (n)
Surgically implanted fish two days postsurgery	38.7 (3.2)	550(144)	25.3 (0.5)	12.9(1.3)	10.1 (0.8)	6.83	9
Control fish two days postsurgery	37.1 (2.7)	518(155)	25.2 (1.4)	12.3(1.8)	7.0 (0.9)	6.83	8
Surgically implanted fish 10 days postsurgery		545(146)	23.6 (0.8)	12.9(1.8)	8.2(1.3)	6.83	9
Control fish 10 days postsurgery		503(151)	23.5(1.3)	11.5 (1.4)	7.2 (0.9)	6.83	8

All experiments were carried out at 9–10°C and all meals consisted of a single prey

Assuming a G. morhua to operate with its individual value, the basic rate parameter ρ_LTE can be considered the outcome of a normally distributed stochastic variable.¹³ Then, the performance of two groups of G. morhua with regard to GER can be compared statistically using a t-test to the values of this parameter calculated for each G. morhua according to Equation (5). Before t-tests were performed (SigmaStat version 3.5) the assumptions of normality (Kolmogorov-Smirnov test) and homogeneous variance of data (Bartlett's test) were tested. Significance was accepted at P < 0.05.

Coefficient of variation (CV) of ρ_LTE is constant and does not depend on the value of this parameter.¹⁵ According to Sokal and Rohlf,¹⁶ this implies that the variances of two values of In ρ_LTE are identical, and that a test of equality of two values of CV is equivalent to H-test of equality of the variances of the logarithmic transforms of ρ_LTE. The latter was therefore used here to test for equality of CV obtained from the different treatments of fish. Significance was accepted at P < 0.05.

Results

Short-term recovery (2 days)

Two days postsurgery, evacuation of force-fed sandeel was considerable slower in fish with surgical implants compared with control fish. The calculated GER parameter ρ_LTE was thus significantly lower (P < 0.001) in implanted G. morhua compared with the control, being respectively 4.2 × 10^-4 (1.2 × 10^-4) and 9.3 × 10^-4 (1.9 × 10^-4) (mean [SD]) (Figure 2). Because the effects of fish length were accounted for by ρ_LTE (Equation (5)), while temperature and energy density of the prey were similar for all animals (Table 1), the observed difference in GERs can be attributed to pure physiological causes. When comparing our data with the value of ρ_LTE = 14.2 × 10^-4 (1.1 × 10^-4) obtained in the study by Andersen¹⁴ on G. morhua feeding voluntarily on A, tobianus (i.e. involving no handling or anaesthesia priory to recovery of the stomach contents), the present GERs only constituted approximately one-third (fish with implants) and two-thirds (control) of this rate (Figure 2), i.e. both being significantly lower (P < 0.001). This difference was also reflected in that on average only 21% and 42% (instrumented and control fish, respectively) of the meals had been evacuated 25 h post-force-feeding, which is inefficient compared with the prediction¹⁴ that 62% of the meal should be evacuated.

Figure 2

Comparison of the basic rate parameter rho (ρ_LTE) two and 10 days postsurgery for force-fed surgically implanted (hatched, n = 9) and control (grey, n = 8) Gadus morhua and with estimates from non-operated voluntarily feeding conspecifics (white).¹⁴ Data are presented as mean (SD), *indicates a significant difference (P < 0.05) between the force-fed groups two days postsurgery, ^†indicates significant difference (P < 0.05) between non-operated voluntarily feeding fish and all force-feeding trials (i.e. both 2 and 10 days postsurgery)

According to Andersen and Beyer¹⁵ the CV of the basic rate parameter ρ_LTE is constant (i.e. independent of the estimated value of this rate parameter). CV should therefore be used when comparing the variability of ρ_LTE in the different groups of G. morhua. For the two groups of force-fed G. morhua, CVs of 0.21 and 0.29 (controls and operated, respectively) were significantly higher (P < 0.002 and P < 0.001, respectively) than the value of 0.08 obtained from G. morhua that fed voluntarily¹⁴ (Table 2).

Table 2

Coefficient of variation (CV) of the basic rate parameter ρ_LTE obtained from G. morhua fed an A. marinus

	Surgically implanted G. morhua (n = 9)	Control G. morhua (n = 8)	G. morhua feeding voluntarily on A. tobianus (n = 20) (from Andersen¹⁴)
Two days postsurgery	0.287^†	0.207^†
Ten days postsurgery	0.242^†	0.348^†	0.080^‡

Different symbols indicate a significant difference

Prolonged recovery (10 days)

Ten days postsurgery, there was no significant difference (P = 0.878) between the rate at which animals with implants evacuated their meal compared with control fish (Figure 2). These estimates of ρ_LTE were however still significantly lower (7.7 × 10^-4 [1.9 × 10^-4] for implanted and 7.9 × 10^-4 [2.8 × 10^-4] for controls) than the value of 14.2 × 10^-4 (1.1 × 10^-4) obtained from G. morhua feeding voluntarily on Ammodytes spp. (P < 0.001).¹⁴ Again, as described above for G. morhua two days postrecovery, the CVs in both of our groups of fish were significantly higher (P < 0.001), being 0.24 in surgically implanted fish and 0.35 in the control group (Table 2) compared with the value of 0.08 obtained from G. morhua that fed voluntarily. In contrast to the improved evacuation ability of the surgically implanted G. morhua following prolonged recovery, the percentage of food that had been evacuated by control G. morhua at c. 23 h post-force-feeding (36%) was comparable both to the week earlier (43% c. 25 h post-force-feeding after 2 days of recovery) and to the value for the surgically instrumented fish (36% c. 23 h post-force-feeding).

Discussion

The present study demonstrates that following two days of postsurgical recovery after introduction of a dummy implant into the body cavity, G. morhua is not capable of evacuating a meal at a rate similar to non-operated fish. By using ρ_LTE (Equation (5)) to compare gastric performance between groups of fish we have accounted for the effects of fish length, while temperature and energy density of the prey were similar for all animals (Table 1). Our results thus suggest that the observed difference in GER can be attributed to pure physiological causes derived either from surgery, postsurgical stress and/or the presence of the implant. But following 10 days of recovery the suppressing effects on gastric performance caused by surgical implantation was gone. Thus at this stage the presence of neither the implant per se nor the surgery was limiting the GER. Reduced postsurgical gastric performance has been reported previously. Twenty-four to 36 h subsequent to invasive surgery, sea bass (Dicentrarchus labrax) L. showed a significantly higher gastric evacuation time (GET) (the relationship between GET and GER can be deduced from Equation (3)) compared with controls, with stomach contents still above 60% of the initial 24 h post-force-feeding, whereas control animals at this time had less than one-third of the meal remaining in the stomach.¹⁷ This is thus in accordance with our results which show that GER had not stabilized in G. morhua two days postsurgery.

A factor that may affect the recovery and even long-term function is the effect on buoyancy. A fish with an implant has to counteract the downward force exerted by the added mass which can be done by secretion of gases into the swimbladder.^18,19 With the present implant in the body cavity, a fish, irrespective of its size, would have to increase its swimbladder volume (5% of the total volume in G. morhua¹⁸) with 14.2 cm³ of air to become neutrally buoyant. In the smallest G. morhua, this implies an almost doubling of swimbladder volume, which may approach the limit for their capabilities.²⁰ Swimbladder adjustment is a low-cost solution but is slow (up to a day or 2),^21,22 meanwhile compensation must occur through active swimming which is energetically expensive.^19,23 Such excess energy use will reduce the aerobic scope available for other processes²⁴ and this may be part of the explanation for the lower gastric performance observed two days postsurgery. Furthermore, active swimming may result in blood being shunted away from the stomach region to prioritize oxygen delivery to the working muscle.^25,26

The use of biotelemetry devices allows experimenters to provide their animals with long recovery periods to facilitate complete recovery from instrumentation procedures. Nonetheless, for biotelemetry systems to hold their true/full potential it is imperative that the fish is capable of dealing with not only the effects arising from anaesthesia, handling and surgical intervention but also any potential disadvantages related to carrying the entire recording/transmitting unit. Taken together, our results show that a potential for returning to presurgical levels indeed exists, as GER was stabilized 10 days postsurgery in fish carrying the implant in the body cavity. Thus the presence of the implant per se did not seem to affect the average gastric performance at this point. The fish in the present study weighed between 290 and 815 g resulting in transmitter:fish mass ratios from 2.3% to 6.5%. It has been suggested that the mass of telemetry tags should not exceed 2% of body mass.^24,27 Several experiments (including experiments on cod) using tags larger than this have however found no significant effect on the swimming performance.^28,29 We found no correlation between GER and body mass indicating that cod can recover from carrying tags that represent up to 6.5% of their body mass. Along the same lines results from salmonids on recovery of swimming performance following surgical implantation of telemetry devices have shown that Juvenile Chinook salmon Oncorhynchus tshawytscha (Walbaum) had a significantly lower critical swimming speed one day postsurgery compared with controls, whereas full recovery was accomplished after 21 days (in-between evaluations are lacking).²¹ Although full recovery was not established, the swimming performance of tagged juvenile Salmo salar L. had improved seven days post-tagging at termination of the experiment.²⁹ Suggestively, the fish may either have recovered in the long term, or the tags, representing on average 8.5% of the fish mass, may have been too heavy or have reduced the mobility of the fish.

When comparing the variability in gastric performance between different groups of G. morhua, the CV of the basic rate parameter ρ_LTE should be used (for further details on this see Materials and methods). The CV value of 0.080 estimated from the voluntarily feeding fish by Andersen¹⁴ (Table 2) compared well with the estimate of 0.098 obtained from a variety of predatory gadoids and their fish prey.¹⁵ This variation probably reflects the interindividual variation in gastric performance. Substantially higher values of CV were obtained from both groups of force-fed G. morhua, which may then be explained by an additional variability due to the suppressing effects of handling, anaesthetization and force-feeding on GER to which the individual fish responds differently. An alike large variability has been observed in O. mykiss; the lag phase from force feeding until the stomach started to empty varied between zero and 5 h.³⁰ As anticipated, for the present G. morhua this furthermore resulted in on average 30% lower values of ρ_LTE from control fish compared with the voluntarily feeding G. morhua in the study by Andersen.¹⁴ Studies on cod using intragastric transmitters have shown that if the fish voluntarily ingest the baited transmitter (in this case a transmitter wrapped in a fillet of herring) high food consumption rates are maintained in the days subsequent to tagging, in contrast to fish tagged by forced insertion (involving handling and anaesthesia) where food intake was notably lower for up to 15 days post-tagging.³¹ This indicates that appetite prevails despite the presence of a transmitter as long as handling and anaesthesia are avoided.

To omit interference from anaesthesia and handling in future studies, one important unanswered question to investigate is how long it takes for instrumented fish to commence voluntarily feeding following the surgical procedure.

Although the primary intention of this study was not to focus on wound healing, this is an important matter, not only for ethical considerations but also because biotelemetric methods enable long-term measurements where open incisions may facilitate internal infections and/or cause changed behaviour and performance thus inflicting data invalidity. Two days postsurgery the surgically implanted fish had no or only slight signs of inflammatory reaction around the incisions. The exception was two fish, one in which the larger wound had opened, and the other where it gapped at one end. These animals were instantly euthanized and data omitted from the analysis. Ten days postsurgery an inflammatory response (redness and slight swelling) was noted around all the larger incision made alongside the basibranchial bone. These observations are in accordance with a previous study in which G. morhua (kept at comparable temperatures of 9.5–14°C) were surgically implanted with dummy transmitters into the body cavity via incision along the linea alba.³² In these animals inflammatory responses begun 5–7 days postsurgery and subsided for 4–8 days with complete wound healing after a total of 24–34 days. Obviously, the time course of, and how to secure proper wound healing following surgical implantation demand further attention.

In summary, two days of postsurgical recovery will not stabilize GER in G. morhua, but 10 days will, when using a standard force-feeding protocol. The results indicate that at 10 days postsurgery, the presence of a dual channel dummy telemetric implant per se did not affect GER but that the effects observed two days after instrumentation are due to surgery, postsurgical stress and/or the presence of the implant. Biotelemetry has the welfare advantages of allowing long recovery periods and avoiding the unnecessary stress arising from handling and confinement, altogether improving quality of data. We believe our results have left a promising future for implementation of fully implantable biotelemetry systems in fish.

References

Fontenot

, Neiffer

DL.

Wound management in teleost fish: biology of the healing process, evaluation, and treatment. Veterinary Clinics of North America. Exo Anim Prac 2004; 7: 57–86

Altimiras

, Larsen

Non-invasive recording of heart rate and ventilation rate in rainbow trout during rest and swimming. Fish go wireless!

J Fish Biol 2000; 57: 197–209

Crocker

, Farrell

, Gamperi

, Cech

JJ.

Cardiorespiratory responses of white sturgeon to environmental hypercapnia. Am J Physiol 2000; 279: R617–28

Gräns

, Olsson

, Pitsillides

, Nelson

, Cech

Jr , Axelsson

Effects of feeding on thermoregulatory behaviours and gut blood flow in white sturgeon (Acipenser transmontanus) using biotelemetry in combination with standard techniques. J Exp Biol 2010; 213: 3198–206

Connors

, Scruton

, Brown

, Mckinley

RS.

The effects of surgically-implanted dummy radio transmitters on the behaviour of wild Atlantic salmon smolts. Hydrobiologia 2002; 483: 231–7

Schulz

UH.

Effects of surgically implanted dummy transmitters on the South American catfish Jundia (Rhamdia quelen). Braz J Biol 2003; 63: 345–8

Knights

, Lasee

BA.

Effects of implanted transmitters on adult bluegills at two temperatures. Trans Am Fish Soc 1996; 125: 440–9

Bauer

, Unfer

, Loupai

Potential problems with external trailing antennas: antenna migration and ingrowth of epithelial tissue - a case study from a recaptured case, Chondrostoma nasus (L.). J Fish Biol 2005; 67: 885–9

Gräns

, Axelsson

, Pitsillides

A fully implantable multi-channel biotelemetry system for measurement of blood flow and temperature: a first evaluation in the green sturgeon. Hydrobiologia 2009; 619: 11–25

10.

Jensen

, Axelsson

, Holmgren

Effects of substrate P and vasoactive intestinal polypeptide on gastrointestinal blood flow in the Atlantic cod Gadus morhua. J Exp Biol 1991; 156: 361–73

11.

Pedersen

, Hislop

JRG.

Seasonal variation in the energy density of fishes in the North Sea. J Fish Biol 2001; 59: 380–9

12.

Jobling

The influences of feeding on the metabolic rate of fishes: a short review. J Fish Biol 1981; 18: 385–400

13.

Andersen

, Beyer

JE.

Mechanistic modeling of gastric evacuation in predatory gadoids applying the square root model to describe surface-dependent evacuation. J Fish Biol 2005; 5: 1391–412

14.

Andersen

NG.

A gastric evacuation model for three predatory gadoids and implications of using pooled field data of stomach contents to estimate food rations. J Fish Biol 2001; 59: 1198–217

15.

Andersen

, Beyer

JE.

Precision of ingestion time and evacuation predictors for individual prey in stomachs of predatory fishes. Fish Res 2008; 92: 11–22

16.

Sokal

, Rohlf

FJ.

Biometry. 2nd edn. San Francisco: W H Freeman, 1981

17.

Axelsson

, Altimiras

, Claireaux

Post-prandial blood flow to the gastrointestinal tract is not compromised during hypoxia in the sea bass Dicentrarchus labrax. J Exp Biol 2002; 205: 2891–6

18.

Harden Jones

, Scholes

Gas secretion and resorption in the swimbladder of the cod Gadus morhua. J Comp Physiol B 1985; 155: 319–31

19.

Strand

, Jørgensen

, Huse

Modelling buoyancy regulation in fishes with swimbladders: biogenetics and behaviour. Ecol Model 2005; 185: 309–27

20.

Tytle

, Blaxter

JHS.

Adaptation by cod and saithe to pressure changes. Neth J Sea Res 1973; 7: 31–45

21.

Adams

, Rondorf

, Evans

SD.

Effects of surgically and gastrically implanted radio transmitters on swimming performance and predator avoidance of juvenile Chinook salmon (Onconhynchus tshawytscha). Can J Fish Aquat Sci 1998; 55: 781–7

22.

Snelderwaard

PCh

, van Ginneken

, Witte

, Voss

, Kramer

Surgical procedures for implanting a radiotelemetry transmitter to monitor ECG, heart rate and body temperature in small Carassius auratus and Carassius auratusgibelio under laboratory conditions. Lab Anim 2006; 40: 465–8

23.

Alexander

RM.

The energetic of vertical migration by fishes. Symp Soc Exp Biol 1972; 26: 273–94

24.

Lafrancois

, Odion

, Claireaux

An experimental and theoretical analysis of the effect of added weight on the energetic and hydrostatic function of the swimbladder of European sea bass (Dicentrarchus labrax). Mar Biol 2001; 139: 13–17

25.

Axelsson

, Fritsche

Effects of exercise and feeding on the gastrointestinal blood flow in the Atlantic cod Gadus morhua. J Exp Biol 1991; 158: 181–98

26.

Dupont-Prinet

, Claireaux

, McKenzie

DJ.

Effects of feeding and hypoxia on cardiac performance and gastrointestinal blood flow during critical speed swimming in the sea bass Dicentrarchus labrax. Comp Biochem Phys A 2009; 154: 233–40

27.

Winter

JD.

Underwater biotelemetry. In: Nielsen

, Johnson

, eds. Fisheries Techniques. Bethesda, MD: American Fisheries Society, 1983: 371–95

28.

Cote

, Moulton

, Frampton

PCB

, Scruton

, McKinley

RS.

Habitat use and early winter movements by juvenile Atlantic cod in a coastal area of Newfoundland. J Fish Biol 2004; 64: 665–79

29.

Lacroix

, Knox

, McCurdy

Effects of implanted dummy acoustic transmitters on juvenile Atlantic salmon. Trans Am Fish Soc 2004; 133: 211–20

30.

Olsson

, Aldman

, Larsson

, Holmgren

Cholecystokinin affects gastric emptying and stomach motility in the rainbow trout Oncorhynchus mykiss. J Exp Biol 1999; 202: 161–70

31.

Winger

, Walsh

SJ.

Tagging of Atlantic cod (Gadus morhua) with intragastric transmitters: effects of forced insertion and voluntary ingestion on retention, food consumption and survival. J Appl Ichthyol 2001; 17: 234–9

32.

Pedersen

, Andersen

NG.

A surgical method for implanting transmitters with sensors into the body cavity of cod (Gadus morhua). Dana 1985; 5: 55–62

Recovery of gastric evacuation rate in Atlantic cod Gadus morhua L surgically implanted with a dummy telemetry device

Abstract

Keywords

Materials and Methods

Fish

Preoperative care

Dummy implant

Surgery

Food

Gastric evacuation experiments

Calculations and statistical analyses

Results

Short-term recovery (2 days)

Prolonged recovery (10 days)

Discussion

References