Introduction

Anaesthetics are widely used in aquaculture to diminish stress and physical injuries during harvesting or transport, and proper anaesthesia is absolutely essential to alleviate pain during surgeries and health examinations (Coyle et al. 2004; Kiessling et al. 2009; Abdolazizi et al. 2011). For aquatic organisms, soluble anaesthetics can be added directly to the water and are readily absorbed across the gills, and transported by the blood to the nervous system; loss of equilibrium and mobility follows rapidly (Coyle et al. 2004; Popovic et al. 2012). However, the anaesthesia also disturbs the respiratory and cardiac function causing hypoxaemia (Soivio et al. 1977; Fredricks et al. 1993; Andersen and Wang 2002). Thus, disturbances of blood gases and haematological parameters occur within seconds of administration, due to splenic contraction and adrenergic activation of the RBC sodium-proton exchanger and hence extracellular acidosis and red cell swelling (Jensen 2004). These respiratory, haematological and metabolic disturbances during anaesthesia have been studied extensively in water-breathing fish, such as Atlantic salmon, rainbow trout, kelp grouper, red drum and perch (Soivio et al. 1977; Iwama et al. 1989; Thomas and Robertson 1991; Iversen et al. 2003; Park et al. 2008; Velíšek et al. 2009), but little is known about such responses in air-breathing fish, which often thrive at higher temperatures than water breathers (Lefevre et al. 2014). Furthermore, it has been suggested that air breathers, with their reduced branchial ion and respiratory gas exchange, are less able to regulate pHe than water breathers (Shartau and Brauner 2014) and, therefore, the restoration of acid–base status and blood gases may be different from those of water breathers.

Striped catfish Pangasianodon hypophthalmus is a tropical facultative air-breathing fish, which is effective at both air and water breathing (Lefevre et al. 2011). Pangasianodon hypophthalmus is intensively cultured in South East Asia and, while various anaesthetics are widely used for transport, there is no information on the rate of recovery or physiological effects. In aquaculture, MS-222 (tricaine methanesulfonate), benzocaine (ethyl para-aminobenzoate), and Aqui-S (50 % isoeugenol) are the most commonly utilised anaesthetics (Coyle et al. 2004; Kiessling et al. 2009; Weber et al. 2009; Abdolazizi et al. 2011). Both benzocaine and MS-222 are local anaesthetics that provide general anaesthesia in fish by blocking voltage-gated sodium channels and hence inhibit neural transmission within the central and peripheral nervous systems (Attili and Hughes 2014). Eugenol is a widely used local analgesic agent to alleviate tooth pain that shares several pharmacological actions with local anaesthetics, including inhibition of voltage-gated sodium channel as well as activation of transient receptor potential vanilloid subtype 1 (TRPV1) (Park et al. 2009). Here, we evaluate the disturbance caused by these three anaesthetics on blood gases and haematological parameters of P. hypophthalmus and the time course for normalisation when allowed to recover in clean normoxic and normocapnic water. In addition, since adrenergic swelling responses have been found in erythrocytes of numerous water-breathing fish (Nikinmaa and Huestis 1984), leading to an increase in cell volume and changes in haematocrit, and since we expected pH and haematocrit changes, we also investigate the existence of this β-adrenergic response in the present study.

Materials and methods

Abbreviations used in this study are shown in Table 1.

Table 1 Abbreviations used in this study
Full size table

Fish. Striped catfish Pangasianodon hypophthalmus weighing 700-1000 g were transferred from local farms to tanks with aerated water at the Department of Aquaculture, Can Tho University (Vietnam) several weeks before the study commenced. During this period, they were fed to satiation with commercial pellets on a daily basis, but fasted for 24 h pre-instrumentation.

Anaesthetics preparation. Benzocaine was pre-dissolved in 3 ml ethanol 70 % and mixed with water (100 mg l−1, Florindo et al. 2006). Aqui-S was dissolved directly in the water at 30 mg l−1 (Iversen et al. 2003). MS-222 was dissolved in 100 mg l−1 tank water with 100 mg l−1 NaHCO3 used as a buffer (Iwama et al. 1989).

Experimental procedures. Each fish was kept in the induction chamber until total loss of equilibrium and reaction to touch (Iwama et al. 1989; Coyle et al. 2004). When anaesthetized, the fish was catheterized into the dorsal aorta using polyethylene tubing (I.D. 0.58 mm, O.D. 0.96 mm) containing heparinized saline (50 IE ml−1) (Soivio et al. 1975), whilst the fish gills were constantly irrigated with aerated water containing one-third of the initial dose of anaesthesia. Times required for induction time and catheterization were recorded for each fish. An arterial blood sample was collected from the catheter immediately after catheterization (0 h) and subsequently at 3, 6, 24, 48, and 72 h of recovery, whilst the fish were maintained in a 200 l tank containing aerated water. Great care was taken not to disturb the fish during recovery. Water temperatures during recoveries were 21.6 ± 0.2, 24.2 ± 0.1, and 25.1 ± 0.2 °C in benzocaine, Aqui-S, and MS-222, respectively.

Blood collection and treatment. Each blood sample (0.3 ml) was collected using a 1 ml syringe, carefully avoiding air bubbles, for measurements of PCO2, pHe, and [lactate] using an iSTAT blood gas analyser (Abbott Laboratories, Abbott Park, Illinois, USA); 0.7 ml of blood was transferred to a 1.5 ml Eppendorf tube and kept on ice for immediate determination of other haematological parameters, such as Hct, [Hb], and RBC counts. The remaining blood was centrifuged at 6000 rpm for 6 min, to separate plasma and RBC, and stored at -80 °C for subsequent analysis of [glucose], [cortisol], [Cl-]e, and [Cl-]i.

Measurement of the haematological and biochemical parameters. Hct was determined by centrifugation in a standard microhaematocrit centrifuge. Hb concentration was determined by the Drabkin’s method; spectrophotometrically at 540 nm using an extinction coefficient of 10.99 mmol−1 cm−1 (Zilstra et al. 1983). Plasma glucose concentration was determined according to Huggett and Nixon (1957), and [cortisol] was determined using a DRG Salivary Cortisol ELISA commercial Kit (USA). RBC counts were determined by counting the number of RBC in a Neubauer chamber under a microscope after diluting 200 times of 0.5 µl blood sample with Natt & Herrick’s stain solution. [Cl-]e was measured using a chloride titrator (Sherwood model 926S MK II chloride analyser). For [Cl-]i, a known mass of RBC pellet was transferred to a known volume of distilled water to induce cell lysis, and Cl- was measured in the haemolysate. The RBC water content was measured gravimetrically in another RBC aliquot from the same centrifuged cell pellet before and after drying at 60 °C for 16 h.

In vitro assessment of red cell adrenergic response. Four fish were catheterized under anaesthesia with benzocaine and allowed to recover for at least 48 h before a 3 ml blood sample was taken. Blood was placed in an Eschweiler tonometer (Kiel, Germany) and equilibrated with humidified gas mixtures supplied from two serially linked Wösthoff gas mixing pumps (Bochum, Germany). Initially, blood was equilibrated with 30 % O2 (PO2 = 216 mmHg) and 3 % CO2 (PCO2 = 21.6 mmHg) to determine blood O2-carrying capacity and then reduced to a PO2 of 15.1 mmHg (approximately 10 % air) at 3 % CO2, resulting in HbO2 saturations of 15-30 %. At this PCO2, pHe is expected to be 7.35 (Damsgaard et al. 2015). The beta-adrenergic agonist isoprenaline was added to the blood to a final concentration of 10−5 mol l−1 (Brauner et al. 2002; Koldkjær et al. 2002). At both steps, the concentration of Hb-bound O2 ([Hb-O2]), Hct, and [Hb] were determined.

Calculations. pHi was calculated from the Donnan-like equilibrium using the ratio of [Cl-]e, [Cl-]i, and pHe (Jensen 2004)

$$pH_{i} = log_{10} \left( {\frac{{\left[ {Cl^{ - } } \right]_{e} }}{{\left[ {Cl^{ - } } \right]_{i} }} \cdot 10^{{ - pH_{e} }} } \right).$$

Based on our previous validation of the iStat for P. hypophthalmus (Damsgaard et al. 2015), the iStat PCO2 was corrected according to the equation:

$$P{\text{CO}}_{2} \left( {\text{true}} \right) \, \left[ {\text{mmHg}} \right] \, = { - }2.17 \, + \, 1.20P_{\text{a}} {\text{CO}}_{2} \;\left( {\text{iStat}} \right).$$

Plasma [HCO -3 ] was calculated from the Henderson Hasselbach equation:

$$\left[ {{\text{HCO}}_{3}^{ - } } \right] = \alpha_{{{\text{CO}}_{2} }} \cdot P{\text{CO}}_{2} \cdot 10^{{{\text{pH}}_{\text{e}} - {\text{pK}}}} ,$$

where αCO2 is the temperature-compensated CO2 solubility in trout plasma (Boutilier et al. 1985) and pK is the pHe-corrected dissociation exponent for CO2 in P. hypophthalmus plasma (Damsgaard et al. 2015).

[Hb-O2] was determined by measuring [O2]total and subtracting physically dissolved O2:

$$\left[ {{\text{Hb}} - {\text{O}}_{2} } \right] = \left[ {{\text{O}}_{2} } \right]_{\text{total}} - \alpha_{{{\text{O}}_{2} }} \cdot P{\text{O}}_{2} ,$$

where [O2]total was determined according to Tucker (1967); αO2 is the temperature-compensated solubility of O2 (Dejours 1981) and PO2 the partial pressure of O2 in the gas mixture.

O2 saturation of Hb (HbO2 sat) during equilibration with 10% air was calculated as

$${\text{Saturation}} = \frac{{\left[ {{\text{Hb}} - {\text{O}}_{2} } \right]_{{10\% {\text{air}}}} }}{{\left[ {{\text{Hb}} - {\text{O}}_{2} } \right]_{{30\% {\text{O}}_{2} }} }}.$$

MCHC was calculated as

$${\text{MCHC}} = \frac{{\left[ {\text{Hb}} \right]}}{\text{Hct}}.$$

Statistics. All data were presented as means ± standard error of the mean. One-way repeated measures ANOVA was applied to determine significant differences amongst the sampling times for each haematological parameter. One sample t test was used to test whether isoprenaline exerted significant effects on the red cells in vitro. A probability (P) value at the 0.05 level was considered as significant.

Results

Time required for induction of anaesthesia, surgery, and recovery stage. A surgical plane of anaesthesia was 3.1 ± 0.2 min for benzocaine, whereas it took 9.3 ± 0.9 and 8.1 ± 0.7 min for MS-222 and Aqui-S, respectively (Fig. 1). Regardless of the anaesthetic, the catheter was inserted and secured within the dorsal aorta in less than 15 min and the fish regained equilibrium within 2-4 min (Fig. 1).

Fig. 1
figure 1

Time recorded for duration of anaesthesia to reach the surgical plane, cannulation, and post-operative recovery of Pangasianodon hypophthalmus with the three anaesthetics. Time values are presented as means ± S.E.M (n = 7). Significant differences between anaesthesia treatments, cannulation, and post-operative recovery are indicated with *, #, and +, respectively

Full size image

Haematological and biochemical parameters. Following full anaesthesia, Hct, RBC counts, and [Hb] changed in a similar manner, with maximal disruption immediately after catheterization, followed by a rapid decrease within 6 h of recovery (Fig. 2a–c). MCHC was stable around 25 mmol l−1 throughout the entire recovery period except with MS-222 (Fig. 2d).

Fig. 2
figure 2

Haematological and biochemical parameters of arterial blood following anaesthesia in Pangasianodon hypophthalmus with Aqui-S (circles), MS-222 (triangles), and benzocaine (crosses) with (a) Hct, (b) RBC counts, (c) [Hb], (d) MCHC, (e) [glucose], (f) [cortisol]. Letters a,m,b indicate significant difference (P < 0.05) from 72 h, for Aqui-S, MS-222, and benzocaine, respectively (one way RM ANOVA). Values are presented as means ± S.E.M (n = 7)

Full size image

All three anaesthetics caused high [glucose] and [cortisol] immediately after surgery. The plasma glucose was high initially within 0-3 h, then decreased gradually during recovery, and normalized at approximately 3 mmol l−1 at 72 h (Fig. 2e). Cortisol concentrations fell by more than a factor of 5 during recovery and stabilized at approximately 30 mmol l−1 within 24 h (benzocaine and MS-222) to 48 h (Aqui-S) (Fig. 2f).

Acid–base status and chloride ions. Immediately after surgery, there was a significant plasma acidosis, which rapidly returned to normal values over the subsequent 6-24 h (Fig. 3a); pHi followed a similar pattern, with a difference between pHe and pHi of around 0.2-0.3 units throughout the experiment for all three anaesthetics (Fig. 3b). Arterial PCO2 was elevated during this time and this elevation was most pronounced in benzocaine (up to 6 mmHg). During recovery from anaesthesia with all three compounds, PCO2 decreased to approximately 2.5 mmHg (Fig. 3c). Similarly, all three anaesthetics caused elevated [lactate], which remained high at 6-8 mmol l−1 immediately after catheterization. In line with PCO2, lactate also recovered almost completely to approximately 0.3 mmol l−1 at 24 h post-operation (Fig. 3d). Immediately after surgery with Aqui-S and MS-222, [HCO -3 ] was depressed, but stabilized within 6 h of recovery at approximately 9 mmol l−1 (Fig. 3e). No significant changes were observed in [Cl-]e and [Cl-]i after anaesthesia, with the concentrations being approximately 100 and 60 mmol l−1, respectively (Figs. 3g, h). Davenport diagrams (Figs. 4a, b, c) show the respiratory status of Pangasianodon hypophthalmus during and after anaesthesia with the three anaesthetics and show that the low pHe immediately after surgery can be largely ascribed to a metabolic acidosis with a minor respiratory component.

Fig. 3
figure 3

Acid–base status and chloride ions of arterial blood following anaesthesia in Pangasianodon hypophthalmus with Aqui-S (circles), MS-222 (triangles), and benzocaine (crosses) with (a) pHe, (b) pHi, (c) PCO2, (d) [lactate], (e) [HCO -3 ], (f) gH2O/g RBC dried weight, (g) [Cl-]e, and (h) [Cl-]i. Letters a,m,b indicate significant difference (P < 0.05) from 72 h for Aqui-S, MS-222, and benzocaine, respectively (one-way RM ANOVA). Values are presented as means ± S.E.M (n = 7)

Full size image
Fig. 4
figure 4

Davenport diagrams of (a) Aqui-S, (b) MS222, and (c) benzocaine with the curved dotted lines indicating PCO2-isopleths and dashed lines indicating in vitro buffer lines taken from Damsgaard et al. (2015). Values are presented as means ± S.E.M (n = 7)

Full size image

Effects of ß-adrenergic stimulation on red cells in vitro. Addition of isoprenaline to whole blood in vitro caused only a minor, albeit statistically significant, rise in blood O2 saturation of 2.8 % (one sample t test, P < 0.01). However, neither Hct nor MCHC was affected (one sample t test, P = 0.1817 and P = 0.1884, respectively) (Table 2).

Table 2 Effect of isoprenaline (10−5 mol l−1) on blood O2 saturation, Hct, and MCHC
Full size table

Discussion

Anaesthesia induction and recovery. The times required to induce anaesthesia and the duration of the subsequent recovery depend on the type, concentration of the anaesthetic, and fish species (da Cunha et al. 2010; Maricchiolo and Genovese 2011). In Pangasianodon hypophthalmus, anaesthesia was reached more rapidly with benzocaine than with Aqui-S and MS-222, whereas recovery was prolonged upon benzocaine anaesthesia compared to the other two anaesthetics. With MS-222 (100-200 mg l−1), silver catfish Rhamdia quelen can be anaesthetized within 1-2.4 min and can recover within 0.25-1.45 min (da Cunha et al. 2010); and with 20–50 mg l−1 Aqui-S, channel catfish Ictalurus punctatus can be anaesthetized within 2–5 min (Stehly and Gingerich 1999). Akbulut et al. (2011) found statistically significant correlations between anaesthetic concentration and recovery time, as well as duration of exposure and recovery time. In general, increased temperature is associated with faster clearance of the anaesthetics, but there was no indication that P. hypophthalmus recovered faster from anaesthesia than water-breathing teleosts studied previously at lower temperatures (Lefevre et al. 2014).

Evaluation of haematological and biochemical parameters. As reported for other species (e.g. Iwama et al. 1989; Iversen et al. 2003; Gholipour et al. 2011), we found that all three anaesthetics caused significant changes in haematological and biochemical parameters of P. hypophthalmus. Most notably, we observed a very pronounced rise in Hct during anaesthesia, and similar albeit typically smaller increases in Hct have been reported in various other species during anaesthesia with MS-222 (Reinitz and Rix 1977; Soivio et al. 1977; Iwama et al. 1989; Molinero and Gonzalez 1995). A rise in Hct can be caused by release of RBCs from the spleen, as a result of plasma loss from the vascular system typically in response to elevated blood pressure, or RBC swelling in response to adrenergic stimulation of the Na+/H+ exchanger (Ferreira et al. 1981; Nikinmaa and Huestis 1984; Wells and Weber 1990; Jensen 2004). In P. hypophthalmus, the rise in Hct was accompanied by proportional increases in [Hb] and RBC counts, such that MCHC did not change. This indicates that swelling of the RBCs did not occur and the present in vitro study showed that the RBCs from P. hypophthalmus do not respond to adrenergic stimulation. The lack of ß-adrenergic activation of the erythrocytes is further substantiated by the similarly low pHi immediately after anaesthesia (Table 2). The present study cannot, however, reveal to what extent the increase of RBC counts is due to haemoconcentration by plasma loss or by splenic release of RBCs. It has been suggested that increasing [Hb] is an adaptation to increase blood oxygen transport capacity during stress (Pereira et al. 2013).

The high [cortisol] and [glucose] observed in P. hypophthalmus in the present study are similar to those reported for water-breathing fish species (Swift 1981; Tomasso et al. 1981; Barton and Peter 1982; Iwama et al. 1989; Ortuño et al. 2002; Park et al. 2008; Velisek et al. 2009; 2011). The mechanisms by which anaesthetics affect cortisol secretion in teleosts are unclear (Iwama et al. 1989; Thomas and Robertson 1991). Molinero and Gonzalez (1995) suggested that anaesthetics act on the hypothalamic–pituitary interrenal (HPI) axis stimulating cortisol secretion. They have also showed cortisol and glucose increases, but such responses were only found at high and intermediate dosages (25 and 30 mg l−1), whereas the lower dose (15 mg l−1) had no significant effect. It can be noted that the fivefold increase in [cortisol] during anaesthesia in the present study agrees with Thomas and Robertson (1991), who saw a similar cortisol increase after restraint and 2 minute air-exposure of red drum Sciaenops ocellatus. Previous studies have suggested that clove oil (isoeugenol) and Aqui-S (50 % isoeugenol) gave rise to lesser elevations in plasma cortisol during light anaesthesia than benzocaine in Atlantic salmon (Iversen et al. 2003) or MS222 in channel catfish (Small 2003). Similarly, isoeugenol had little effect on plasma cortisol in rainbow trout (Wagner et al. 2003).

The acid–base status during anaesthesia and recovery. While a large body of literature exists on acid–base and ion regulation in a variety of water-breathing fish species, little is known concerning these regulatory processes in bimodal breathers (Shartau and Brauner 2014). Acid–base disturbance in fish blood, resulting in increased plasma PCO2 and a reduction in pHe, can be induced by environmental challenges including as hypoxia and hypercapnia, as well as by exhaustive exercise (Baker et al. 2009; Shartau and Brauner 2014). Breathing air is associated with elevated PCO2 as a result of the large difference in the solubility of CO2 in air and water (Dejours 1981). Further air breathers are often poor at pHe regulation, possibly because of their reduced gills and reduced branchial irrigation leading to the suggestion that there is a compromise between oxygen uptake and ion and pH regulation in air-breathing fish (Ishimatsu and Itazawa 1983; Shartau and Brauner 2014). The present study with P. hypophthalmus anaesthesia reveals rapid pHe regulation, which is thus very unusual in air-breathing fish.

The cessation of ventilation during anaesthesia results in significant hypoxaemia and/or hypercapnia and is associated with respiratory acidosis (Iwama et al. 1989; Cooper and Morris 1998; Andersen and Wang 2002). This was also consistent with our findings in P. hypophthalmus, where a significant increase of PCO2 immediately after catheterization contributed to the acidosis. The rise in PCO2 is indicative of impaired gas exchange during anaesthesia despite the gills being constantly irrigated with aerated water during surgery. The marked reduction in pHe was, however, primarily due to the production of lactic acid, i.e. metabolic in origin, presumably in response to severe hypoxaemia in various tissues, caused by a probable reduction in blood flows. The almost 20-fold rise of lactate subsided within the first few hours during recovery and was attended by normalization of pHe and plasma [HCO -3 ] as well as reduction in PCO2, presumably as normal ventilation of both gills and swim bladder were re-established. The increase of [lactate] in P. hypophthalmus is consistent with other studies on the effects of anaesthesia in water-breathing fish such as Atlantic salmon Salmo salar, brook trout Salvelinus fontinalis, and rainbow trout Salmo gairdneri (Houston et al. 1971; Soivio et al. 1977; Olsen et al. 1995; Iversen et al. 2003). Such increases in [lactate] are normally seen in the acid–base disturbance in fish after anaerobic exercise or following exposure to hypoxic conditions (Wood et al. 1977).

Post-anaesthesia recovery of blood gases has been well studied in water-breathing fish (Soivio et al. 1977; Molinero and Gonzalez 1995; Cooper and Morris 1998); whereas very little is known about recovery in these parameters in air-breathing fish. It is known that gills play a central role in acid–base regulation in water-breathing fish, which accounts for approximately 90 % of total ion transport during regulation from an acid–base disturbance (Claiborne et al. 2002; Evans et al. 2005). However, gill surface areas of bimodal air breathers in general are reduced (Hughes and Morgan 1973), which may prolong the acid–base recovery time. It has been suggested that P. hypophthalmus is unusual among air-breathing fish in having very large and well-developed gills at the same time as an air-breathing organ and that it can cover its entire oxygen requirements through the water phase in normoxic water (Lefevre et al. 2011). The present study supports this finding in that the rapid regulation of the acidosis induced by anaesthesia is regulated with rapidity, reminiscent of an active water breather such as rainbow trout.

Conclusions

All three anaesthetics effectively immobilized the fish and reduced responses to tactile stimulation to a level where transport or minor surgical procedures could be performed, but they also caused significant changes with similar patterns on gas and haematological parameters which were generally normalized within 24 h. The increase in Hct and [Hb] of Pangasianodon hypophthalmus caused by the anaesthetics resulted probably from increased RBC numbers and there was no indication of RBC swelling, which is different from active water-breathing fish such as rainbow trout. In addition, P. hypophthalmus is unusual among air-breathing fish with its strong capacity for acid–base regulation.