Introduction

Hemorrhagic shock as a trauma-related condition can be one of the major causes of morbidity and mortality in veterinary critical care patients (Ko et al., 2012). Rapid fluid therapy with crystalloids or colloids remains the routine intervention in resuscitation of patients suffering hemorrhagic  shock  (Gao et al., 2018). However, subsequent monitor- ing to evaluate the efficacy of resuscitation  is vital, because excessive fluid therapy or uncorrected hypovolemia can lead to delete- rious effects (Ko et al., 2012; Marik, 2010). Therefore, understanding the tools and methods accessible for intravascular volume assessment is critical.

Thoracic radiography is one of the most important and frequently performed diag- nostic modalities to evaluate heart diseases and has the potential to provide information equivalent to other cardiac diagnostic mo- dalities (Gugjoo et al., 2013). Also, it is com- monly employed to assess volume status and serial thoracic radiography may be helpful in determining effects of fluid therapy (Kalan- tari et al., 2013). Vertebral heart score (VHS) is an accessible, applicable, and interpretable diagnostic method for objectively evaluating cardiac size on radiographs as well as for the sequential assessment of cardiac size on se- rial radiographs of the same patient (Gugjoo et al., 2013; Johnson et al., 2008).

Recently, many studies have been carried out to determine the normal values of VHS  in different breeds of dogs such as Yorkshire Terrier, Boston Terrier (Jepsen-Grant et al., 2013), Belgian Malinois (Almeida et al., 2015), Indian Spitz (Bodh et al., 2016), Ger- man Shepherd, and Iranian native dogs (Gh- adiri et al., 2010). Although VHS changes in hypovolemic and hypervolemic conditions have been documented, to our knowledge its

variation during fluid therapy in hemorrhag- ic shock condition has not been investigated previously. Therefore, the objective of the study reported here was to assess breed-spe- cific VHS as a novel parameter for endpoint determination in fluid  resuscitation  phases of dogs with severe hemorrhagic shock us- ing lactated Ringer’s solution and 6% Hy- droxyethyl starch.

Materials and Methods

Animals: The present experimental study was approved by Animal Care and Research Committee of Shahid Chamran  University of Ahvaz. Ten male native dogs, 1.5 to 3.5 years old and weighing 18.56 ± 4.80 kg were used. They were in good body condition and considered healthy by physical examination, electrocardiography, echocardiography, and some  hemato-biochemical parameters.

Animals Preparation and Instrumenta-  tion: After cannulation of right cephalic vein with an 18-gauge catheter (for drugs and fluid administration), the anesthesia was induced with an intravenous bolus dose of propofol (Lipuro 1%, Braun, Melsungen, Germany) (6.0 mg/kg) and fentanyl (Cas- pian Tamin, Rasht, Iran) (5 μg/kg) (Braz et al.,  2004).  Followed  by  intubation  with an

8.0 to 8.5 mm cuffed endotracheal tube the animals were placed in right lateral recum- bency and anesthesia was maintained with isoflurane (1.8%) in 100% oxygen  (Ko  et al., 2012). The right femoral artery was ex- posed by sterile technique and cannulated with a 16-gauge catheter that was connected to a 3-way stopcock for induction of bleed- ing and to measure arterial blood pressure. Body temperature was maintained at 37° to 38°C with a heating blanket (Nascimento et al., 2006). Also, an experienced  investigator

evaluated physical parameters including mu- cous membrane color, capillary refill time and peripheral pulse quality routinely at the end of each step.

Experimental Protocol: Assessment of VHS was performed at the end of each step (A1 to A8) in five distinct stages as follows: Control stage (A1): Baseline assessment was  obtained  after  induction  of  anesthesia

and instrumentation.

Hemorrhagic stage (A2): Each dog was then hemorrhaged to a mean arterial pressure (MAP) ranging from 40 to 50 mmHg (Braz et al., 2004). The procedure lasted approxi- mately 30 min and blood was collected into sterile empty blood bags.

Hypovolemic stage (A3): The animals were left in shock situation for an addition-  al 30 min period during which no fluid was administered. If a physiologic compensatory mechanism developed and MAP increased above the purpose values, more blood was removed to restore the MAP back to 40 to  50 mmHg.

Resuscitation stage (A4 – A7): Animals were then randomly allocated into two equal groups. Group A was resuscitated with lac- tated Ringer’s solution (LR) (Iranian Paren- teral and Pharmaceutical Co., Tehran,  Iran) at 20 ml/kg in 15 min for four consecutive times. Group B was resuscitated with 6% Hydroxyethyl Starch (HES) (Voluven, Fre- senius Kabi, Homburg, Germany) at 5 ml/kg in a similar manner to group A.

Post-resuscitation stage (A8): The last assessment was performed one hour after termination of resuscitation. Then, animals were allowed to recover from  anesthesia.

Radiographic assessment: Right lateral recumbent thoracic radiographs of each dog were obtained using a digital radiography (DR) system (Aero DR detector, Konica-Mi-

nolta, Japan, 35 mAs and 65 kVp). VHS was determined by a veterinary radiologist based on the previously described method (Bu- chanan and Bücheler, 1995). Briefly, the car- diac long axis (LA) was measured from the ventral border of the left main stem bronchus to the most distant point of the apex. The cardiac short axis (SA) was measured per- pendicular to the long axis at the level of the caudal vena cava. The measurements of the two axes were positioned over the thoracic vertebra, starting at the cranial edge of the 4th thoracic vertebra (T4)  and the number  of vertebra (v) spanned was estimated to the nearest 0.1 vertebral body length. The VHS was calculated from the sum of the short and long axes.

Statistical analysis: Data of this study were evaluated by repeated measures anal- ysis of variance and LSD post hoc using SPSS 16.0 statistical software (SPSS, Inc., Chicago, USA). All data were presented as mean ± standard deviation (SD). A p value of less than 0.05 was considered as statisti- cally significant.

Results

The  average  blood  volume  loss  was  54

± 4 ml/kg, which corresponded to approxi- mately 62% ± 4% of the estimated circulat- ing blood volume (88 ml/kg); so that there were no significant differences between the groups (p>0.05). In both groups, the ani- mals responded to fluid therapy and were successfully resuscitated with no mortality. Dogs in group A received 386 ± 102.85 ml LR, whereas dogs in group B received 89.5 ±

22.17 ml HES at each step of fluid resuscita- tion. There was no significant difference be- tween the studied groups according to MAP during all study stages (p>0.05).

Figure 1 shows VHS variation during fluid

Figure1. Mean ± SD values of VHS in dogs with experimen- tal hemorrhagic shock after resuscitation with LR and HES.

therapy in dogs with hemorrhagic shock. The mean ± SD of VHS in both groups before in- duction of shock was 9.3 ± 0.59, which was significantly reduced in A2, as shown in Fig. 2. Following the fluid resuscitation, this val- ue increased and approximately returned to pre-shock values at steps 6 and 7  in groups A and B, respectively. It is worth noting, the statistical  analysis  indicated  no   significant

difference between solution types  (p>0.05).

Discussion

In this experimental model of hemorrhagic shock, the short-term effects of LR and 6% HES, as the most frequently used crystalloid and synthetic colloid fluids, on the VHS were studied from the clinical point of view.

Although there are many subjective and semiquantitative methods of assessing heart size, the VHS is the preferred method at present (Woolley et al., 2007). It gained pop- ularity among veterinary radiologists due to various advantages such as justifiable ana- tomic markers for measurement of cardiac silhouette, good correlation with both echo- cardiographic and electrocardiographic pa- rameters, and is independent of the observer experience (Almeida et al., 2015; Gugjoo et al., 2013).

It is very important that the breed and re- cumbent side of radiographic  view  should be taken into consideration while calculating VHS in dogs to have a high specificity for normal heart size (Bodh et al., 2016). In our country, native dogs account for a large num- ber of companion animals. Although they are

Figure 2. Demonstration of VHS in control (A) and hemorrhagic (B) stages (9.6v and 8v respectively).

not a registered breed, Ghadiri et al. (2010) established normal VHS values  for  Irani-  an native dogs with mean of 9.4 vertebras. Our results at the control stage also confirm this finding. It is proposed that the cardiac silhouette may have a different shape on a right lateral versus a left lateral radiographic view (Olive et al., 2015). But interestingly, according to the study by Ghadiri and his colleagues (2010), no significant difference was found between the VHS of right  later-  al and left lateral radiographs in native dogs. Furthermore, in the present study, we used right lateral radiographs in order to obtain the most accurate information about cardiac size (Greco et al., 2008).

The heart is an organ in which notable changes in intravascular volume can affect  its size. Wilson et al. (2010) demonstrated that severely anemic cats have high VHS values, which can be due to volume over- load caused by hemodynamic compensatory mechanisms. In another study it was docu- mented  that  primary   hypoadrenocorticism in dogs can lead  to a significant  reduction  in VHS, that may be attributed to hypovo- lemia (Melián et al., 1999). Normal mean VHS value in dogs is 9.7 v, and values be- tween 8.7 and 10.7 v are generally consid- ered to be physiological (Kraetschmer et al., 2008). However, in our investigation, induc- tion of hemorrhagic shock led to significant decrease  in VHS values to a mean of 7.7 v  in the studied groups. This phenomenon can be explained by the severe hypovolemia that results in decline of filling pressure with sub- sequent decrease in the diameters of the car- diac chambers and the fiber length of cardiac muscles (Traber et al., 1993).

Restoration of intravascular volume is the first therapeutic goal in hemorrhagic shock (Bouglé  et al.,  2013). The amount  of   fluid

required to restore intravascular volume is very variable and so frequent assessment is needed. Also, it is crucial during the resus- citation phase to assess  whether  the patient is fluid-responsive or not (Kalantari et al., 2013). In the current study, following admin- istration of intravenous fluids, the VHS val- ues increased slightly and almost returned to the control values that can indicate success- ful resuscitation. Hence, assessment of the VHS can be a reasonable method to monitor the response to fluid resuscitation. The lack  of significant difference in VHS values be- tween the two types of solutions that were used, despite different dosage, probably is due to the discrepancy in mechanisms of ac- tion as well as the stability time in the vessels (Nascimento et al., 2006).

Because of the similarity of our recorded VHS values in the control stage with stan- dard values reported for native dogs by Gh- adiri et al., (2010), utilization  of this  meth- od in breeds of dogs with a predetermined standard VHS can be valuable  in  estimat- ing resuscitation endpoint. However, some studies mentioned that noncardiac factors such as body condition score (Jepsen-Grant  et al., 2013), recumbency side (Greco et al., 2008), training (Zahabpour et al., 2016), and interobserver variability (Olive et al., 2015) can affect VHS. Therefore, to increase the reliability of this method in determination of resuscitation endpoint, it is necessary to es- tablish a universal protocol.

Conclusion

The findings of this study indicate that VHS assessment can guide clinicians in the estimation of volume status in dogs with hemorrhagic shock as well as monitoring the response to fluid  resuscitation.  Moreover,  in daily  practice,  serial  assessment  of VHS