Shock is commonly diagnosed in both human and veterinary emergency rooms and is defined as inadequate cellular energy production due to impaired tissue perfusion, which results in insufficient oxygen delivery (DO2) to meet oxygen demand.1–3 There are many classifications of shock; however, hypovolemic shock due to hemorrhage is the most commonly diagnosed in human emergency rooms.3
Prompt recognition of hemorrhagic shock is imperative to guide therapy and prevent mortality in the emergency setting. By simplifying the definition of shock to a state where DO2 does not meet tissue oxygen demand, direct measurement of cardiac output (CO) with a pulmonary arterial catheter would be ideal to diagnose hemorrhagic shock. However, this is not accessible in most facilities nor clinically feasible in critically ill patients given its invasive nature. Without this direct measurement, clinicians rely on a combination of history and evaluation of perfusion parameters, including mentation, heart rate, pulse quality, mucous membrane color, capillary refill time, and core-toe-web temperature gradient, to diagnose shock. Low blood pressure is also commonly used to support a diagnosis of shock. However, perfusion parameters and blood pressure may be minimally affected in hypovolemic shock due to neurohormonal compensatory processes that can result in increased CO, heart rate, and systemic vascular resistance. Shock index, the ratio of heart rate to systolic blood pressure, has also been used as a marker of shock in human and veterinary patients, with recent studies4–10 citing a shock index of > 1 as an accurate marker of shock when compared to controls. In addition to physical examination, diagnostic tests including lactate and base excess, may be used to support a diagnosis of shock. Recently, studies11 have shown that the caudal vena cava collapsibility measured using point-of-care ultrasound can be used clinically to evaluate volume status and resuscitation in shock.
Near-infrared spectroscopy (NIRS) measures oxygenated and deoxygenated hemoglobin in tissues, reflecting the balance of tissue DO2 and consumption.8 In humans, this tool has been shown to predict the need for blood product administration in trauma victims that otherwise appear cardiovascularly stable.12 Multiple veterinary studies have published data evaluating the use of this technology in dogs, establishing reference intervals in healthy dogs,13 comparing tissue oxygen saturation (StO2) in dogs on room air versus nasal oxygen supplementation,14 and documenting the positive correlation between StO2 and DO2 in hemorrhagic shock models and naturally occurring shock patients.15–17 To date, no studies have been performed evaluating the correlation of StO2, DO2, and CO.
The purpose of this study was to determine if StO2 was correlated with DO2 and CO and could therefore be used as a marker of shock in dogs. The second purpose was to determine if StO2 could be used as a marker of resuscitation. We hypothesized that StO2 would be positively correlated with DO2 and CO and could therefore be used as a marker of shock and resuscitation.
Methods
Animals
Eight purpose-bred spayed female Beagles were used in this study. Each dog was determined to be healthy based on physical examination, complete blood count, and biochemistry panel. All procedures were approved by the IACUC at Colorado State University. This study was performed from September to December 2021.
Anesthesia and instrumentation
Dogs were premedicated with hydromorphone (0.1 mg/kg, IM), and an IV catheter was placed in a cephalic vein. Following preoxygenation, general anesthesia was induced with propofol (5 to 10 mg/kg, IV) to effect. A cuffed endotracheal tube was placed and connected to an anesthetic circle system, and general anesthesia was maintained with isoflurane vaporized in 100% fraction of inspired oxygen. Dogs were spontaneously breathing throughout the study.
Dogs were instrumented with ECG, pulse oximetry, invasive blood pressure monitoring, esophageal temperature probe, and end-tidal CO2 via sidestream sampling for the duration of general anesthesia. Normothermia was maintained using a forced-air warming device and heated water blankets.
After induction and placement of anesthetic monitoring, the StO2 probe was placed in a clipped 5 X 5 cm region over the sartorius muscle on the upward-facing leg. A 20-gauge over-the-needle catheter was placed in a dorsal metatarsal artery, and a 16-gauge over-the-needle catheter was placed in a jugular vein. A 7.5F Swan-Ganz catheter (Biosensors International Inc, Japan, Tokyo) was placed using a flow-directed technique via the jugular vein, and the location in the pulmonary artery was confirmed via pressure waveform analysis. All catheters were placed using aseptic technique.
StO2, CO, and DO2 measurements
Measurement of StO2 was performed using a noninvasive portable near-infrared spectroscope (ODISsey Tissue Oximeter, ViOptix Inc), which has been validated in animal models.18,19 Prior to data collection in each dog, the machine was calibrated per manufacturer instructions. Per manufacturer recommendations, any StO2 measurement with a signal quality < 80% was not reported.
Measurement of CO via thermodilution was performed per manufacturer guidelines. A temperature probe was placed in an ice bath of syringes. A 5-mL bolus of 0.9% NaCl was injected into the proximal port of the Swan-Ganz catheter, and CO was calculated by system software (Hemosphere Advanced Monitoring Platform, Edwards LifeSciences). All CO measurements were obtained in duplicate with 3 to 5 minutes between measurements, and the average of measurements was used for statistical analysis.
After completion of the study, DO2 was calculated using the following equation: DO2 = CO X {(1.34 X Hb X SpO2) + (0.003 X PaO2)}, where SpO2 is peripheral oxygen saturation of hemoglobin as measured by pulse oximetry and PaO2 is alveolar partial pressure of oxygen.
Experimental design
Once each dog was anesthetized and instrumented, the mean arterial pressure (MAP) was stabilized at 70 to 80 mm Hg for 10 minutes before time point (TP)-1 measurements were obtained. If necessary, dobutamine (0.5 to 3 μg/kg/min, IV), phenylephrine (0.5 to 2 μg/kg/min, IV), and/or an IV bolus of crystalloids (3 to 20 mL/kg) were used to stabilize blood pressure within the desired range. Two patients received an intervention (dopamine, dobutamine) to achieve target blood pressure for TP1. These were discontinued prior to blood removal.
A low CO state was induced by removing the dog's blood from the jugular catheter over 20 minutes, targeting a MAP of 40 mm Hg or until 60% of the dog's total blood volume (blood volume, 90 mL/kg) was removed, whichever occurred first. Once a MAP of 40 mm Hg was reached, blood withdrawal was stopped. Blood was stored in blood collection bags containing citrate phosphate dextrose adenine. After 10 minutes, TP2 measurements were collected. If MAP had increased from 40 mm Hg during the 10-minute wait time, no further blood was withdrawn. Fifty percent of the withdrawn blood was returned back to the dog over 15 minutes, and 10 minutes later, TP3 measurements were collected. The remaining shed blood was returned over an additional 15 minutes, and after 10 minutes, the final TP4 measurements were obtained. After completion of the study, dogs were deinstrumented and allowed to recover from general anesthesia. All dogs survived without complication and participated in further research for a limited time prior to being adopted.
CO, arterial blood pressure, heart rate, SpO2, body temperature, StO2, and end-tidal CO2 were recorded at all TPs. An arterial blood gas was performed at all TPs. Hemoglobin and PaO2 from the arterial blood gases were used to calculate DO2.
Statistical methods
Descriptive statistics were compiled for each variable and timepoint. In order to investigate changes over time, a mixed model was fit separately for each response variable. Specifically, TP1–4 were included as a fixed effect. Dogs were included as a random effect to account for the repeated measures design. The Dunnett method was used to compare downstream TPs to baseline (TP1). Residual diagnostic plots were used to evaluate model assumptions of normality and equal variance. For CO and DO2, log transformation was used to satisfy model assumptions. Normally distributed variables are reported as mean (± SD), and non-normally distributed variables are reported as median (range). Correlations between variables were calculated accounting for repeated measures on subjects.20 Statistical analyses were performed by commercial software (SAS Institute Inc). A value of P < .05 was considered statistical evidence of a difference.
Results
Eight healthy female spayed purpose-bred Beagles were enrolled in this study. The mean age was 4.8 years (± 1.0). The mean body weight was 8.8 kg (± 1.56).
The mean blood volume withdrawn was 36.3 mL/kg (± 11.6). The MAP at TP1, TP2, TP3, and TP4 was 74.1 mm Hg (± 7.4), 49.5 mm Hg (± 13.7), 63.5 mm Hg (± 13.3), and 71.4 mm Hg (± 8.88), respectively.
The median CO at TP1 was 2.7 L/min (range, 1.2 to 3.8). This decreased significantly at TP2 (0.6 L/min; range, 0.5 to 1.5; P < .001 vs TP1; Tables 1 and 2). The median CO was 2.0 L/min at TP3 (range, 1 to 2.6; P = .098 vs TP1; Tables 1 and 2) and 2.5 L/min at TP4 (range, 1.8 to 5.3; P = .709 vs TP1; Tables 1 and 2).
Median [range] or mean (SD) values for cardiovascular variables of tissue oxygen saturation (StO2), global oxygen delivery (DO2), cardiac output (CO), blood removed, peripheral oxygen saturation of hemoglobin as measured by pulse oximetry (SpO2), mean arterial pressure (MAP), and hemoglobin concentrations (Hb) at time points (TPs).
| Variable | TP1 | TP2 | TP3 | TP4 |
|---|---|---|---|---|
| StO2 (%) | 77 (± 9.54) | 44.8 (± 19.5) | 63.1 (± 9.85) | 82.5 (± 12.6) |
| DO2 (mL/min/m2) | 460.7 [227.7–686.9] | 121.6 [82.1–246.6] | 311.1 [194.7–415.0] | 373.7 [266.1–908.0] |
| CO (L/min) | 2.7 [1.2–3.8] | 0.6 [0.5–1.5] | 2.0 [1.0–2.6] | 2.5 [1.8–5.3] |
| Blood removed (mL/kg) | 0 | 36.3 (± 11.6) | 18.4 (± 5.98) | 0 |
| SpO2 | 96.0 (± 2.62) | 95.9 (± 2.53) | 96.9 (± 1.69) | 96.5 (± 1.66) |
| MAP (mm Hg) | 74.1 (± 7.4) | 49.5 (± 13.7) | 63.5 (± 13.3) | 71.4 (± 8.88) |
| Hb (g/dL) | 13.5 (± 2.03) | 13.6 (± 2.18) | 11.9 (± 1.75) | 11.1 (± 1.63) |
Statistical comparison (P value) between cardiovascular findings at TP1 versus individual TPs.
| Variable | TP1 vs TP2 | TP1 vs TP3 | TP1 vs TP4 |
|---|---|---|---|
| StO2 | < .001 | .002 | .47 |
| DO2 | < .001 | .047 | .98 |
| CO | < .001 | .098 | .71 |
Bolded values are statistically significant.
CO = Cardiac output. DO2 = Global oxygen delivery. StO2 = Tissue oxygen saturation. TP1 = Time point (TP)-1 baseline. TP2 = Maximum blood removed. TP3 = 50% shed blood returned. TP4 = 100% shed blood returned.
The median DO2 at TP1 was 460.7 mL/min/m2 (range, 227.7 to 686.9). This decreased significantly at TP2 to 121.7 mL/min/m2 (range, 82.1 to 246.6) when compared to TP1 (P < .001; Tables 1 and 2; Figure 1). Median DO2 was significantly different at TP3 at 311.1 mL/min/m2 (range, 194.8 to 415) when compared to TP1 (P = .047; Figure 1). Median DO2 was not significantly different at TP4 (373.7 mL/min/m2 [range, 266.1 to 908]) compared to TP1 (P = .98; Figure 1). Due to machine error in reading hemoglobin, 1 dog did not have a calculable DO2 at TP3 and was therefore left out of statistical analysis regarding DO2.
Median and range oxygen delivery (DO2) values at time point (TP)-1 (baseline), TP2 (maximum blood removed), TP3 (50% shed blood returned), and TP4 (100% shed blood returned). P values comparing TP1 versus TP2, TP3, and TP4 are < 0.001, 0.047, and 0.98, respectively (Table 1). Asterisks represent statistically significant differences between TPs.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.23.12.0297
The mean StO2 at TP1 was 77.8% (± 9.54). This decreased significantly at TP2 to 44.8% (± 19.5) when compared to TP1 (P < .001; Tables 1 and 2; Figure 2). Mean StO2 increased to 63.1% (± 9.85) at TP3 but remained significantly lower compared to TP1 (P = .002). There was no difference between the mean StO2 at TP1 and TP4 (82.5% ± 12.6; P = .466).
Median and range tissue oxygen saturation (StO2) values at TP1 (baseline), TP2 (maximum blood removed), TP3 (50% shed blood returned), and TP4 (100% shed blood returned). P values comparing TP1 versus TP2, TP3, and TP4 are < 0.001, 0.002, and 0.47, respectively (Table 1). Asterisks represent statistically significant differences between timepoints.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.23.12.0297
There was a strong positive correlation between StO2 and CO (r = 0.80; P < .001; Figure 3) and a strong positive correlation between StO2 and DO2 (r = 0.75; P < .001; Figure 4).
Repeated measures correlation curves between tissue oxygen saturation (StO2) and cardiac output. Each line style and symbol pair indicate the results for 1 dog. The overall correlation is r = 0.80; P < .001.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.23.12.0297
Repeated measures correlation curves between tissue oxygen saturation (StO2) and global oxygen delivery (DO2). Each line style and symbol pair indicate the results for 1 dog. The overall correlation is r = 0.75; P < .001.
Citation: American Journal of Veterinary Research 85, 7; 10.2460/ajvr.23.12.0297
Discussion
This study's hemorrhagic shock and resuscitation model in healthy dogs found strong positive correlations between StO2 and DO2 and StO2 and CO, supporting our hypothesis. StO2, as measured by a NIRS device, offers an objective, noninvasive, portable, easy-to-use method to assess patients at presentation and during resuscitation.13,15 The median StO2 decreased significantly after removal of the maximal amount of blood in this study (TP2), suggesting that StO2 can be used in conjunction with physical examination and diagnostic parameters as a noninvasive tool to support a diagnosis of shock. The StO2 increased after 50% of the shed blood was returned to the patient and increased further when 100% of the shed blood was returned, suggesting that StO2 can be used as a marker of resuscitation in hemorrhagic shock. The baseline (TP1) StO2 was comparable to prior studies in dogs with and without supplemental oxygen.13,14
While not a focus of this study, CO at all TPs was reflective of acute blood loss and replenishment back to baseline after resuscitation. Baseline CO values were within normal reference ranges reported in a veterinary study examining CO in a hemorrhagic shock model.21 CO, as measured by thermodilution, was used as the gold standard in this study for determining shock and resuscitation. However, it requires placement of a pulmonary arterial catheter, which has been associated with increased mortality or morbidity in human studies, raising concern for its safety and efficacy in critically ill patients.22
The positive correlation between StO2 and DO2 in this study is consistent with a prior study17 with a similar design but using a different NIRS machine. The prior study17 documented a stronger correlation between StO2 and DO2 (r = 0.97; P = .005), which could be due to a larger sample size of 14 Beagles compared to 8 in the present study or due to the use of a different NIRS machine. The present study included TP3 after 50% of shed blood was returned, which allowed for documentation of trends in CO, StO2, and DO2 at multiple levels of hypovolemia, which was not included in the previous study.
Despite withdrawing blood until the target MAP of 40 mm Hg was reached, the mean MAP at TP2 was 49.5 mm Hg. The discrepancy between the target MAP and TP2 MAP is likely due to activation of endogenous compensatory mechanisms, such as activation of the renin-angiotensin-aldosterone system, fluid shifts, baroreceptor activation, and peripheral vasoconstriction during the 10-minute wait time between cessation of blood withdraw and TP2 measurements.
The limitations of this study include the small sample size of 8 dogs. Additionally, only female dogs were used in this study. The deleterious effects of extrapolating data from one sex to another has been highlighted in human literature, citing sex as a biologic variable.23 A recent study24 documented significant differences in stroke volume index and CO between healthy males and females at rest using different measures of echocardiography. There are currently no studies comparing CO between male and female dogs. These were all clinically normal dogs prior to undergoing general anesthesia. Blood loss in this model was performed under controlled conditions while receiving 100% inhaled fraction of inspired oxygen, which is not representative of the clinical presentation of hemorrhagic shock. Additionally, this model did not induce trauma or tissue damage, which is commonly associated with hemorrhagic shock. Trauma and tissue damage may induce a significant inflammatory response, which may impact hemodynamic coherence.25 Inhaled anesthetics will contribute to some level of cardiovascular disruption. In addition, the absence of conscious fear, pain, and/or anxiety is eliminated under general anesthesia, which could impact heart rate, CO, and subsequently DO2 by nature of the equation. Individual variance in pigmentation and body condition score may contribute to the wide range of StO2 readings.13 In our study, the NIRS probe was placed on an area of white skin in all dogs.
In conclusion, in this canine fixed-pressure hemorrhagic shock model, there was a strong positive correlation between StO2 and DO2 and between StO2 and CO. There was a significant decrease in StO2 after the withdrawal of blood, suggesting that it may be used in conjunction with physical examination and diagnostic parameters to support a diagnosis of shock. In addition, StO2 returned to baseline values after the return of all shed blood, suggesting that it may be used as a marker of resuscitation in hemorrhagic shock. Further studies in larger groups and clinical patients are warranted.
Acknowledgments
This study was presented at the 24th Annual Colorado State University College of Veterinary Medicine and Biomedical Science Research Symposium 2023, Fort Collins, Colorado.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
This study was completed at the C. Wayne McIlwraith Translational Medicine Institute, Colorado State University, Fort Collins, Colorado, and was supported by a College Research Council grant.
References
- 1.↑
de Laforcade A, Silverstein DC. Shock. In: Silverstein DC, Hopper K, eds. Small Animal Critical Care Medicine. 2nd ed. Saunders Elsevier; 2015:26–30.
- 2.
Cooper E. Pathophysiology of shock. In: Drobatz K, Hopper K, Rozanski EA, Silverstein DC, eds. Textbook of Small Animal Emergency Medicine. 1st ed. John Wiley & Sons; 2019:981–985.
- 3.↑
Richards JB, Wilcox SR. Diagnosis and management of shock in the emergency department. Emerg Med Pract. 2014;16(3):1–22.
- 4.↑
Koch E, Lovett S, Nghiem T, Riggs RA, Rech MA. Shock index in the emergency department: utility and limitations. Open Access Emerg Med. 2019;11:179–199. doi:10.2147/OAEM.S178358
- 5.
Peterson KL, Hardy BT, Hall K. Assessment of shock index in healthy dogs and dogs in hemorrhagic shock. J Vet Emerg Crit Care. 2013;23(5):545–550. doi:10.1111/vec.12090
- 6.
Porter AE, Rozanski EA, Sharpe CR, Dixon KL, Price LL, Shaw SP. Evaluation of the shock index in dogs presenting as emergencies. J Vet Emerg Crit Care. 2013;23(5):538–544. doi:10.1111/vec.12076
- 7.
Kraenzlin MN, Cortes Y, Fettig PK, Bailey DB. Shock index is associated with mortality in canine vehicular trauma patients. J Vet Emerg Crit Care. 2020;30(6):706–711. doi:10.1111/vec.13013
- 8.↑
Middleton P, Davies S. Noninvasive hemodynamic monitoring in the emergency department. Curr Opin Crit Care. 2011;17(4):342–350. doi:10.1097/MCC.0b013e328348bf9b
- 9.
Talbot CT, Zersen KM, Hess AM, Hall KE. Shock index is positively correlated with acute blood loss and negatively correlated with cardiac output in a canine hemorrhagic shock model. J Am Vet Med Assoc. 2023;261(6):874–880. doi:10.2460/javma.22.11.0521
- 10.↑
Boyd C, Smart L. Hypovolemic shock. In: Drobatz K, Hopper K, Rozanski EA, Silverstein DC, eds. Textbook of Small Animal Emergency Medicine. 1st ed. John Wiley & Sons; 2019:986–992.
- 11.↑
Lisciandro GR. POCUS: Global FAST – Rapidly detecting treatable forms of shock, advanced life support, and cardiopulmonary resuscitation. In: Lisciandro GR, ed. Point-of-Care Ultrasound Techniques for the Small Animal Practitioner. John Wiley & Sons; 2021:729–755.
- 12.↑
Beekley AC, Martin MJ, Nelson T, et al. Continuous noninvasive tissue oximatery in the early evaluation of the combat casualty: a prospective study. J Trauma Acute Care Surg. 2010;69(suppl 1):S14–S25. doi:10.1097/TA.0b013e3181e42326
- 13.↑
Hall KE, Powell L, Beilman G, Shafer KR, Skala VK, Olmstead EA. Measurement of tissue oxygen saturation levels using a portable near-infrared spectroscopy in clinically healthy dogs. J Vet Emerg Crit Care. 2008;18(6):594–600. doi:10.1111/j.1476-4431.2008.00369.x
- 14.↑
Sullivan LA, Campbell VL, Radecki SV, Webb CB. Comparison of tissue oxygen saturation in ovariohysterectomized dogs recovering on room air versus nasal oxygen insufflation. J Vet Emerg Crit Care. 2011;21(6):633–638. doi:10.1111/j.1476-4431.2011.00693.x
- 15.↑
Berg AN, Conzemius MG, Evans RB, Tart KM. Evaluation of tissue oxygen saturation in naturally occurring canine shock patients. J Vet Emerg Crit Care. 2019;29(2):149–153. doi:10.1111/vec.12818
- 16.
Pavlisko ND, Henao-Guerrero N, Killos MP, et al. Evaluation of tissue oxygenation saturation with near-infrared spectroscopy during experimental acute hemorrhagic shock and resuscitation in dogs. Am J Vet Res. 2014;75(1):48–53. doi:10.2460/ajvr.75.1.48
- 17.↑
Pavlisko ND, Killos M, Henao-Guerrero N, Ricco CH, Were S. Evaluation of tissue hemoglobin saturation (StO2) using near-infrared spectroscopy during hypoxemia and hyperoxemia in Beagle dogs. Vet Anaesth Analg. 2016;43(1):18–26. doi:10.1111/vaa.12258
- 18.↑
Bello G, Ahad A, Mao JM. Oxygen saturation in arterial and venous occlusions for an animal flap model. Proc SPIE. 2007;6424:642401–642411.
- 19.↑
Mao J, Xu R, Lash B, et al. Quantitative evaluation of ViOptix's tissue oximeter in an ex-vivo animal model. Proc SPIE. 2008;6849(68490C):2–11.
- 20.↑
Bland JM, Altman DG. Calculating correlation coefficients with repeated observations: part 1–Correlation within subjects. BMJ. 1995;310(6977):446. doi:10.1136/bmj.310.6977.446
- 21.↑
Shih A, Giguère S, Vigani A, Shih R, Thuramalla N, Bandt C. Determination of cardiac output by ultrasound velocity dilution in normovolemia and hypovolemia in dogs. Vet Anaesth Analg. 2011;38(4):279–285. doi:10.1111/j.1467-2995.2011.00604.x
- 22.↑
Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators. JAMA. 1996;276(11):889–897. doi:10.1001/jama.1996.03540110043030
- 23.↑
Lee SK. Sex as an important biological variable in biomedical research. BMB Rep. 2018;51(4):167–173. doi:10.5483/BMBRep.2018.51.4.034
- 24.↑
Patel HN, Miyoshi T, Addetia K, et al; WASE Investigators. Normal values of cardiac output and stroke volume according to measurement technique, age, sex, and ethnicity: results of the world Alliance of societies of echocardiography study. J Am Soc Echocardiogr. 2021;34(10):1077–1085.e1.
- 25.↑
Kleinveld DJB, Hamada SR, Sandroni C. Trauma-induced coagulopathy. Intensive Care Med. 2022;48(11):1642–1645. doi:10.1007/s00134-022-06834-7
