| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL ARTICLE |
Department of Cardiovascular Surgery, Türkiye Yüksek Ihtisas Hospital, Ankara
1 Department of Surgery, University Hospital, Uppsala, Sweden
2 Department of Anesthesiology, Atatürk Training and Research Hospital for Chest Disease and Thoracic Surgery, Ankara, Turkey
A Tulga Ulus, MD, Tel: +90 532 522 15 20, Fax: +90 312 229 01 48, Email: uluss{at}yahoo.com, Department of Cardiovascular Surgery, Türkiye Yüksek Ihtisas Hospital, 06100 Sihhiye, Ankara, Turkey.
ABSTRACT
To analyze the effect of central venous pressure on cerebrospinal fluid oxygen tension and intrathecal pressure, multiparameter sensors were introduced into the intrathecal space for continuous monitoring of cerebrospinal fluid PO2, PCO2, and intrathecal pressure in 15 pigs. After 20 min of aortic clamping, hypervolemia was established for 20 min, followed by normovolemia. The animals were divided into 3 groups: in group 1, cerebrospinal fluid PO2 = 0% at some time during crossclamping; in group 2, cerebrospinal fluid PO2 was <50%; and in group 3, cerebrospinal fluid PO2 remained 
50%. Mean decreases in cerebrospinal fluid PO2 during the initial 20 min of crossclamping were 82%, 57%, and 15% in groups 1, 2, and 3, respectively. Following induction of hypervolemia, central venous and cerebrospinal fluid pressures increased simultaneously; this caused a significant decrease in cerebrospinal fluid PO2 in group 2 only. In this model, aortic clamping did not increase cerebrospinal fluid pressure if central venous pressure was not elevated. The detrimental effect of elevated intrathecal pressure on cerebrospinal fluid oxygenation was seen only in animals with an intermediate degree of spinal cord ischemia. This might have important implications for the prevention of paraplegia during thoracoabdominal aortic replacement.
Key Words: Central Venous Pressure • Cerebrospinal Fluid Pressure • Spinal Cord Ischemia
INTRODUCTION
Experimentally, drainage of cerebrospinal fluid (CSF) has consistently reduced the occurrence of neurologic deficits due to thoracic aortic occlusion. It has been shown that central venous pressure (CVP) increases after aortic crossclamping, which might be related to blood from the heart being pumped predominantly to the upper torso, with an increase in upper body blood volume.1 It has been speculated that the rise in CVP might result in increased CSF pressure and thus have a detrimental effect on spinal cord perfusion during aortic crossclamping.1 The aim of this study was to assess the relationship between CVP and CSF oxygen tension in a pig model when CVP was deliberately increased by hypervolemic hemodilution during aortic crossclamping.
MATERIAL AND METHODS
All animals were treated in compliance with the "Principles of Laboratory Animal Care" and the "Guide for the Care and Use of Laboratory Animals" (National Institutes of Health Publication No. 80–23, revised 1985). Study approval was obtained from the local ethics committee for animals. Fifteen pigs, weighing a mean of 25.4 ± 0.7 kg, were anesthetized with morphine 20 mg and pancuronium bromide 8 mg intravenously after premedication with tiletamine 3 mg · kg–1, zolazepam (Zoletil 100, Reading, France) 3 mg · kg–1, and xylazine (Rompun Vet, Bayer, Germany) 2.2 mg · kg–1. Each animal was connected to a respirator (Servo 900D; Siemens Elema, Sweden) with a tidal volume of 10 mL · kg–1. Ventilation was maintained with 40% O2 and air mixture to assure oxygenation with PaO2 of 14–22 kPa. Continuous intravenous anesthesia was used with an infusion of glucose 25 mg · mL–1, ketamine 5 g (Ketaminol Vet, 100 mg · mL–1), morphine 120 mg, and pancuronium bromide (Pavulon) 60 mg at a rate of 4 mL · kg–1. Normothermia was maintained with warm fluids and a thermostatically controlled heating pad. The right carotid and femoral arteries were catheterized (18 gauge catheter; Ohmeda Medical Devices, Murray Hill, NJ, USA) for arterial pressure monitoring. Central venous pressure was also monitored via the right jugular vein. All hemodynamic measurements were recorded using an Ohmeda transducer connected to a Siemens Sirecus 1280/1281 surveillance monitor. A median sternotomy was performed, and the dissection was extended to the abdominal aorta and the left common iliac artery through a left paramedian retroperitoneal incision. Both subclavian arteries and the entire descending thoracic and abdominal aorta together with the iliac arteries and the abdominal visceral, intercostal (T4–13), and lumbar arteries were dissected. A limited thoracic laminectomy was performed, and a multiparameter PO2, PCO2, pH, and temperature sensor (Paratrend 7; Biomedical Sensors, High Wycombe, UK) was introduced over an arterial needle introducer into the intrathecal space for continuous CSF monitoring of PO2, PCO2, and pH, as previously described.2,3 The probe consisted of a combined electrode-fiberoptic system that simultaneously measured PO2, PCO2, pH, and temperature after a 30-min calibration period. A laser Doppler probe (T201; Perimed, Sweden) was placed posterior to the dura (cranial to the level of Paratrend catheter insertion) and connected to a laser Doppler flowmeter (Pf2B; Perimed, Sweden) for measurement of spinal cord blood flow. The thoracic aorta was clamped just distal to the left subclavian artery, and a second crossclamp was placed above the truncus coeliacus. The animals were divided into 3 groups retrospectively, based on spinal cord oxygen tension: in group 1 (n = 3), CSF PO2 reached 0% at some time during crossclamping; in group 2 (n = 6), PO2 fell to <50% of the baseline value; and in group 3 (n = 6), PO2 remained 50% of baseline.
All variables were recorded at baseline and after 18 mL · kg–1 blood was taken from the animals by infusing 12 mL · kg–1 Ringer’s lactate solution and 10 mL · kg–1 Macrodex within 30–45 min. Partial exsanguination was achieved without any change in hemodynamic parameters, creating normovolemic hemodilution. Ringer’s lactate and Macrodex were used to replace the blood volume lost because of the wide dissection during this period. After 5 min, both aortic crossclamps were placed. The 1st 20 min normovolemic ischemic period was followed by a 2nd 20-min period of hypervolemia obtained by infusion of 60 mL · kg–1 Macrodex and the previously removed blood. In the 3rd 20-min period, normovolemia was restored by withdrawing blood (Figure 1
). Paratrend and laser Doppler recordings were continuously obtained during each period. All animals were scarified at the end of the experiment.
|
RESULTS
CSF pressure and CVP responded similarly in all stages and reached a maximum of 25 mm Hg and 20 mm Hg, respectively, in the 2nd period of the study (Figures 2 and 3). The ranges of mean CSF pressure and CVP in each 20-min period are given in Figure 1
; both pressures were significantly increased above the baseline in the 2nd period. CSF oxygen tension decreased by 82.3%, 56.5%, and 15.2% in group 1, 2, and 3, respectively (p < 0.05) during the 1st 20 min period (Figure 4). In the next 20 min with high CVP and CSF pressure, PO2 decreased to 90.3%, 85.1%, and 33.5% in group 1, 2, and 3, respectively; only in group 2 was the decrease significant compared to the 1st 20 min. In the last 20 min of the experiment, oxygen tensions were not significantly different to those in the previous stage. CSF PCO2 began to increase in the 1st 20 min, the increment was accelerated in the 2nd period, and settled in the 3rd period in each group (Figure 5). CSF pH reacted like CSF CO2, with decreases in the 2nd period settling in the 3rd period; there were no significant differences among groups (Figure 6). Spinal cord blood flow decreased throughout the 2nd period and reached 54.2%, 65.1%, and 79.9% of baseline at the 20th min of the 2nd period in groups 1, 2, and 3, respectively. It increased in the final period to 66.7%, 100.6%, and 86.6% of baseline in groups 1, 2, and 3, respectively; there were no significant differences among groups (Figure 7). During the partial exsanguination, mean hematocrit decreased to 29.7% ± 12.2% (12.1%–57.0%). The mean spinal cord oxygen tension increased from 7.7 ± 2.4 to 7.8 ± 2.5 kPa (p> 0.05), pH decreased from 6.64 ± 0.32 to 6.61 ± 0.35 (p> 0.05), PCO2 decreased from 6.9 ± 0.9 to 6.4 ± 0.8 kPa (p < 0.05), and spinal cord blood flow increased by 35.4% ± 18.2% (p < 0.05) during normovolemic hemodilution. Most hemodynamic variables and spinal cord temperature did not differ significantly among groups (Table 1).
|
|
|
|
|
|
|
To reach to a high CVP for a spinal cord ischemia study, we designed a hypervolemic experimental model that started with normovolemic hemodilution. Although normovolemic hemodilution was not the primary focus of this study, the results from this period provide a secondary endpoint of the experiment. The hemodynamic changes that occur during normovolemic hemodilution have been well characterized, but its effect on spinal cord oxygenation is unclear. Acute limited normovolemic hemodilution is safe and efficacious. Its advantage is the potential improvement in tissue perfusion associated with decreased viscosity.4 Despite reduced oxygen-carrying capacity after cute limited normovolemic hemodilution, tissue oxygenation is achieved by an augmented oxygen extraction ratio. Nevertheless, Wisselink and colleagues5 found that spinal cord injury was worsened by normovolemic hemodilution in their model. In our study, following exsanguination, hemoglobin values decreased by approximately 30%. In this stage, although CSF PCO2 decreased (6%) and PO2 increased (4%) insignificantly, tissue blood flow increased significantly (35%). Accordingly, acute limited normovolemic hemodilution may be used as a therapeutic tool for oxygenation of tissues during spinal cord ischemia.
Paraplegia can result from inadequate blood flow to the spinal cord, and this is an inevitable consequence of occlusion of the aorta. CSF drainage may increase the perfusion pressure of the spinal cord and hence reduce the risk of ischemic spinal cord injury. Although this treatment was first suggested 30 years ago, concern about possible complications of CSF drainage has prevented this treatment from gaining wide clinical acceptance. On the other hand, this complication is undoubtedly multifactorial; therefore, drainage of CSF may not protect all patients from paraplegia.6
The importance of CVP and CSF pressure changes resulting from thoracic aortic occlusion and their relationship to the development of spinal cord ischemia has not been clarified yet. There are many controversial studies concerning CVP, CSF pressure, and the benefits of CSF drainage after aortic crossclamping. According to Aadahl and colleagues,7 CVP did not increase; whereas Drenger and colleagues8 found that CSF pressure increased, but CVP did not. Ryan and colleagues9 could not find any relationship between CSF pressure and CVP after thoracic aortic crossclamping; but others showed a strong correlation of CVP with CSF pressure. In this study, CVP and CSF pressure did not increase following aortic crossclamping, but they reacted together with a strong correlation. The study design, with retrospective grouping and a relatively small number of studied animals, is the major limitation of our study.
We also investigated the effect of altering CSF pressure on spinal cord oxygenation and CVP. To clarify this, many animal models of thoracic aortic crossclamping have been developed, although some closely resemble the clinical situation, others do not. The results of our previous studies demonstrated that our model of aortic crossclamping produces immediate alterations in CSF oxygenation, which correlate well with the microcirculatory changes measured by epidural laser Doppler flowmetry.2,3 After 20 min of aortic clamping, we increased CVP; in this period, CSF pressure behaved similarly to CVP and both reached approximately 20 mm Hg. Spinal cord oxygenation was not affected in this stage in groups 1 and 3, but it decreased significantly in group 2. This indicates that if CSF pressure increases, the partially ischemic animals (group 2) are most effected.
Drainage of CSF has uniformly lowered the incidence of paraplegia in different experimental models. Miyamoto and colleagues10 found that paraplegia was prevented almost completely by drainage of CSF with a 60-min aortic occlusion. Blaisdell and Cooley11 showed that the incidence of paraplegia decreased from 50% to 8% when CSF pressure was lowered in dogs. On the other hand, Svensson and colleagues12 were unable to demonstrate a beneficial effect of CSF drainage in baboons. Wadouh and colleagues13 found that CSF drainage in pigs offered no protection against spinal cord damage. Piano and Gewertz14 suggest that the role of CSF drainage in preventing spinal cord ischemia may be quite variable. The protective effect of CSF drainage was also found to be limited by Kazama and colleagues.15 In our study, increased CSF pressure did not decrease the spinal cord oxygenation in all animals; it affected mostly the intermediately ischemic group. According to a review by Khan and Stansby16 of randomized trials involving CSF drainage during aortic surgery, there are limited data supporting the role of CSF drainage in prevention of neurological injury in aortic surgery. Although clinical trials using this intervention have met with success, a prospective trial by Crawford and colleagues17 showed no benefit. Dasmahapatra and colleagues18 concluded that CSF pressure is an important factor in determining the degree of spinal cord ischemia during aortic occlusion, and suggested that measures to reduce CSF pressure would mitigate the degree of spinal cord ischemia. Kazama and colleagues15 concluded that removal of CSF offers spinal cord protection only when CSF pressure is abnormally elevated. More recently, Safi and colleagues19 presented their experience of 1,004 cases of thoracic and thoracoabdominal aortic aneurysm surgery with and without CSF drainage: in type II aneurysms, a neurological deficit was determined 6.6% of patients who had CSF drainage and in 29% of those without CSF drainage. Long-term survival also improved with CSF drainage, and they concluded that CSF drainage is safe and effective for reducing morbidity and mortality. A recent prospective randomized trial, focusing only on CSF drainage during repair of type I and II thoracoabdominal aortic aneurysms, indicated an 80% reduction in the relative risk of paraplegia and paraparesis in patients who had CSF drainage.20
Although spinal cord ischemia was obtained in the same experimental setting in all groups, the degree of spinal cord oxygenation differed, probably because of unpredictable collateral circulation towards the spinal cord. It was concluded that aortic crossclamping does not cause an increase in CSF pressure if CVP is not elevated. The detrimental effect of elevated intrathecal pressure on CSF oxygenation was seen only in animals with an intermediate degree of spinal cord ischemia. These findings might have important implications for the prevention of paraplegia during thoracoabdominal aortic replacement.
Presented at the 14th Annual Meeting of the Mediterranean Association of Cardiology and Cardiac Surgery, Bodrum, Turkey, September 26–29, 2004.
REFERENCES
Asian Cardiovasc Thorac Ann 2009;
17:46-53
© 2009 by SAGE Publications
DOI: 10.1177/0218492309102534
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
