Raised intracranial pressure (ICP) for any reason is a life-threatening problem and reducing the partial pressure of carbon dioxide (PaCO2) is one of the most effective ways of decreasing it . Conventional lung ventilatory support of patients with raised ICP consists of positive-pressure ventilation with either volume control (VC) or pressure control (PC), and volume-controlled ventilation with positive end-expiratory pressure (PEEP) . Pressure-controlled inverse ratio ventilation (PC-IRV) of the lungs - where the duration of inspiration exceeds that of expiration - has also been investigated in patients with raised ICP and several mechanisms may influence the ICP in these circumstances. First, the change of ventilatory mode may alter PaCO2. Second, intrathoracic pressure changes may be transmitted to the intracranial space via the venous system or through the thoracic intervertebral foramina . Third, with intact cerebral autoregulation, lowering of arterial pressure causes a rise in ICP. All these mechanisms may induce changes in ICP . Likewise, Clarke , reported that PC-IRV had a minimal net effect on ICP when used in patients with an already raised ICP; it was concluded that changes in PaCO2, rather than in mean airway pressure (PAW), were responsible.
The present study investigated the effects of two modes of inverse ratio ventilation: (1) pressure- and (2) volume-controlled (VC-IRV) on ICP, cerebral perfusion pressure (CPP), and haemodynamics in rabbits with raised ICP due to induced subarachnoid haemorrhage (SAH). Normocapnia was maintained throughout the experiments.
Our accredited Animal Care Committee approved all the procedures undertaken. Twenty-four New Zealand white rabbits of either gender weighing between 3300 and 3620 g were used. They were fasted overnight, but allowed water.
Surgery and anaesthesia
Anaesthesia was induced, with the rabbits in a Plexiglas® box, with 4% isoflurane in 100% oxygen. The right marginal ear vein was cannulated and an infusion of 0.9% NaCl solution was started at 10 mL kg−1 h−1. A tracheotomy was performed and a cannula (Portex tracheal tube 3 mm i.d., Portex-SIMS, Portex Ltd, Hythe, UK) was inserted into the trachea. Vecuronium 0.1 mg kg−1 and fentanyl 1 μg kg−1 were given. The lungs were ventilated with isoflurane 2.2% in air and oxygen (FiO2 =0.5) using intermittent positive pressure ventilation (IPPV) from a lung ventilator (Evita 2®; Dräger, Lübeck, Germany) - which was adjusted to maintain normocapnia - at a respiration rate of 20 breaths min−1 and tidal volume of 8 mL kg−1. Further doses of vecuronium (0.1 mg kg−1) and fentanyl (1 μg -kg−1) were given as necessary. PETCO2 was continuously measured with a capnograph (Anesthesia Gas Monitoring 1304®; Bruel & Kjaer, Copenhagen, Denmark) and maintained in the range 4.7-5.3 kPa. End-tidal isoflurane concentrations were adjusted to 1.0-1.2 minimal alveolar concentration. The left ear artery was cannulated to permit monitoring of arterial pressure and blood-gas status. Arterial pressure was monitored throughout (Hellige SMU 612 PBG®; Hellige GmbH, Freiburg, Germany). A 20-G catheter (Braun Cavafix-MT®; Braun Melsungen AG, Melsungen, Germany) was introduced into the right external jugular vein to measure central venous pressure (CVP). Body temperature was maintained at 38-39°C using a heat lamp controlled by a servomechanism sensed by a temperature probe in the nasopharynx.
The animals were operated upon in the supine position on a surgical table. The skin was infiltrated with bupivacaine 0.25%, and a 20 mm incision was made over the left parietal region. A left parietal burrhole (10 mm diameter) was then opened just anterior to the left coronary suture and an epidural probe (Probe 1®; Spiegelberg, Hamburg, Germany) was placed for monitoring ICP. The space between the probe and the edge of the burrhole was filled with acrylic to eliminate any possible pressure gradient.
The occipitoatlantal ligament was exposed, through a median nuchal skin incision with the underlying muscles split, in the rabbits in Groups 3 and 4 in the prone position. Autologous blood (4 mL) was injected over 10 s through a 22-G needle into the cisterna magna while the ICP was monitored continuously . The animals were positioned head-down for 20 min to facilitate the settling of blood in the basal cisterns, as described by Arthur and colleagues . The skin wound was then closed with 4-0 silk sutures. No intervention was made for the next 30 min in order to permit stabilization.
Baseline measurements were obtained from the time intermittent positive-pressure ventilation of the lungs was instituted. After SAH had been induced (Groups 3 and 4), the ventilatory mode was changed to PC-IRV in Groups 1 and 3 and to VC-IRV in Groups 2 and 4. Measurements were then determined at 5, 15 and 30 min:
• Group 1 (control, n = 6): PC-IRV (I:E = 2:1) was maintained for 30 min.
• Group 2 (control, n = 6): VC-IRV (I:E = 2:1) was maintained for 30 min.
• Group 3 (SAH, n = 6): PC-IRV (I:E = 2:1) was maintained in animals with settled SAH for 30 min.
• Group 4 (SAH, n = 6): VC-IRV (I:E = 2:1) was maintained in animals with settled SAH for 30 min.
During PC-IRV, I:E = 2:1 was applied at a ventilatory rate of 18-22 breaths min−1; in order to achieve a square-wave flow pattern, the peak inspiratory pressure was 19-22 cm H2O, and PaCO2 was maintained between 4.7 and 5.3 kPa.
During VC-IRV, an I:E ratio = 2:1 was applied at a ventilatory rate of 18-22 breaths min−1; in order to achieve a square wave flow pattern, the tidal volume was adjusted to between 20 and 22 mL and PaCO2 maintained between 4.7 and 5.3 kPa.
Monitoring of physiological variables
Heart rate, MAP, CVP and nasopharyngeal temperature were recorded from the monitor. PaCO2, partial pressure of oxygen (PaO2), and pH were analysed from arterial blood samples; haematocrit (Hct) and plasma sodium concentration were measured from the venous blood samples. These physiological variables were determined at the outset, and then at 5, 15 and 30 min from the start of the experiments.
Monitoring of ICP, CPP and PAW
The epidural probe was connected to a cerebral pressure monitor (Brain-Pressure Monitor®; Spiegelberg, Hamburg, Germany) for the continuous measurement of ICP. Not only the alterations in ICP, but also the changes in CPP were calculated and recorded automatically by the monitor from the following equation: CPP = MAP − ICP.
PAW was measured throughout the experiments via the lung ventilator. These pressure recordings were obtained at the outset and then at 5, 15 and 30 min from the commencement of the experiments.
Statistical analysis was carried out using SPSS® for Windows v. 8.0 (SSPS, Inc., Chicago, IL, USA). Data from all groups were expressed as means ± standard deviation. All variables were analysed using non-parametric testing: physiological variables were interpreted using the Kruskal-Wallis test followed by the U-test when indicated. The Friedman analysis of variance was used for within-group comparisons, and if this showed a difference, the source of the difference was identified by comparing the groups using a Wilcoxon signed rank sum test. P < 0.05 was considered as statistically significant.
Heart rate, CVP, Hct, pH, PaCO2, serum sodium concentration and body temperature did not change over time in any of the groups (Table 1). Only PaO2 increased gradually during the experimental period in each group (P < 0.05) (Table 2). However, no significant difference was observed during intergroup comparisons. On the other hand, MAP decreased during the experimental period in each group (P < 0.02). MAP initially reached to higher values in raised ICP periods with SAH than those without SAH (P < 0.02), but changes in MAP did not appear to be affected by the ventilation mode intergroup comparisons.
Changes in ICP, CPP and PAW
As soon as SAH was induced, ICP measurements in Groups 3 and 4 became significantly higher than those in the controls (P < 0.02) (Table 2). Moreover, ICP increased during the experimental period as compared with baseline in each group (P < 0.05), However, ICP measurements did not seem to be affected by the ventilation modes. CPP decreased during the experimental period compared with base-line in each group (P < 0.05), but neither SAH nor ventilation mode seemed to affect CPP significantly. PAW increased during the experimental period compared with baseline in each group (P < 0.02) and it was raised more by VC-IRV than by PC-IRV (P < 0.02).
Many factors interfere with cerebral blood flow and ICP. These include systemic arterial pressure, Hct, body temperature, PaCO2, PaO2 and anaesthetic agents . In our study, body temperature, Hct and PaCO2 remained constant. We tried to maintain PaCO2 in the range 4.7-5.3 kPa. We preferred to use isoflurane in our experiments because it is the volatile anaesthetic agent that least affects cerebral autoregulation and it does not alter CBF provided the dose is kept below 1.5 MAC [9,10].
Increased PAW due to the use of inverse ratio ventilation of the lungs, leads to an elevation of intrathoracic pressure and the prolonged inspiratory period also results in trapping of air in the lungs and rise in intrinsic PEEP [11,12]. The use of a decelerating inspiratory flow pattern in PC-IRV is characterized by a rapid increase in inspiratory pressure, as well as in PAW, since most of the total volume is given during the beginning of the inspiratory period . In our study, we also observed an increase in PAW of each group during either PC-IRV or VC-IRV, which was statistically remarkable in the latter. Conversely, PC-IRV has been found to induce a greater PAW compared with volume-controlled ventilation in a number of studies [12-16]. In those studies, exponentially decelerating flow patterns - which was used during pressure controlled ventilation - should be mentioned as a differential variable. However, we could only apply a constant flow during either of the ventilatory modes because our device could not be varied.
The fact that different modes of lung ventilation changed ICP significantly was another common finding in all four groups of our study and this finding may raise the question whether increased PAW could be responsible or not. Previous studies have often investigated the effects of PEEP on ICP. Among the few studies concerning the effects of PC-IRV on ICP, one was derived from nine patients with head injury who were treated in an intensive care unit . In the first part of that study, patients' lungs were ventilated by application of pressure control with I:E ratios of 2:1 for an hour; tidal volume was kept constant and although significant differences in intrinsic PEEP, peak and PAW were observed no changes were found in ICP, end-tidal CO2, MAP or CPP. Clarke  attributed the alterations of ICP, which ranged from −6 to +5 mmHg, to changes in PaCO2 ranging from −0.3 to +0.4 kPa. In the second part of that study, end-tidal CO2 was constant. As some alterations were observed in tidal volume, peak and mean PAWs, and intrinsic PEEP, no significant change was found in ICP. They concluded that alterations in ICP during inverse ratio ventilation were due to the changes in PaCO2.
Feldman and colleagues , who assessed the effects of PEEP on intracranial compliance in rabbits, found that a PEEP of 10 cmH2O reduced intracranial compliance. Raised intrathoracic and CVPs, decreased return of venous blood to the heart, and depressed cardiac output were held responsible for shifting the ICP-volume curve to the left after PEEP was applied. Cerebral blood and cerebrospinal fluid begin to be displaced when an increase in intracranial volume develops. However, when intrathoracic pressure increases, flow through the jugular veins decreases and hence drainage of cerebral venous blood becomes a problem. The result is a rise in either CBF or ICP, which could be avoided, to some extent, by elevation of the head. The increase in PAW and associated decrease in CPP can also be considered as other mechanisms that reduce intracranial compensatory reserves during lung ventilation with PEEP. In another study assessing the ICP-volume relationship and compliance in patients with head injury, Burchiel and colleagues  applied a PEEP of 5-30 cm H2O to four groups of patients. They observed that ICP was raised significantly in patients with disturbed pressure-volume relationships after the use of PEEP, and cardiac output was markedly decreased whenever PEEP was >15 cm H2O. Interestingly, ICP remained unchanged during lung ventilation with PEEP in their two patients whose intracranial compliances had already been disturbed. Their coexisting severe pulmonary diseases, which caused reduced pulmonary compliance, were believed to protect against a rise in ICP because the increased end-expiratory pressure was not transmitted. The negative effects of PEEP on cerebral blood flow and ICP, which were frequently observed in patients requiring PEEP, led us to investigate whether the use of inverse ratio ventilation would be better than ventilation with PEEP in case of need. However, inverse ratio ventilation - applied as either pressure- or volume-controlled ventilation - might have exerted a negative effect on ICP while producing a rise in intrathoracic pressure. Besides, our device was unable to generate a decelerating inspiratory flow pattern, so that increases in PAW might have changed ICP significantly.
When different modes of lung ventilation are evaluated, PAW is vital in determining arterial oxygenation and oxygen delivery . Abraham and Yoshihara  found that pressure-controlled ventilation enhanced arterial oxygenation to a greater extent than volume-controlled ventilation, and they suggested that increased intrinsic PEEP and PAW, as well as better alveolar ventilation and distribution of inspired gas, were responsible for this superiority. Contrary to this, we found that increases in PAW were greater, and arterial oxygenation was better, during volume-controlled IRV. The fact that arterial oxygenation increased less with volume-controlled IRV - although MAP increased during PC-IRV - might be due either to a lesser rise in PAW or to alterations in ventilation/perfusion ratio, or both.
Although MAP influences ICP markedly when the cerebral autoregulatory response is disturbed, the effects of inverse ratio ventilation on haemodynamics are still being debated. During inverse ratio ventilation, haemodynamics and cardiac output are indirectly related to MAP. Despite studies suggesting no effect , there are numerous reports supporting a disturbing effect on haemodynamics [11,13,15,16,19]. Mang and colleagues  found that PC- or VC-IRV produced similar changes in arterial oxygenation, lung ventilation, cardiac output and haemodynamics in a sheep model. They concluded that PAW remained unchanged and the level of applied PEEP always exceeded the inflection point on the inspiratory pressure-volume curve. Abraham and Yoshihara  demonstrated that no significant changes had occurred in cardiorespiratory variables after 30 min when PC-IRV (with an I:E ratio = 2:1) was used. In those two studies [12,19], PC-IRV was compared with volume-controlled ventilation with PEEP and increases in PAW, as well as depression in cardiac output, were observed in all. They concluded that positive inotropic agents would be able to compensate for the resultant low mixed venous oxygen saturation. In another study, Ludwigs and colleagues  investigated the effects of volume-controlled ventilation alone, volume-controlled ventilation with PEEP, VC-IRV and pressure-controlled IRV in pigs. Except for volume-controlled ventilation, all methods reduced cardiac output and PC-IRV seemed the worst. Because the highest PAW was reached during PC-IRV, this mode was held responsible for the highest tendency to produce haemodynamic changes and they suggested this would result in a lower delivery of oxygen to peripheral tissues. In our experiments, increase in PAW during both PC- and VC-IRV might have led to a rise in intrathoracic pressure, which tended to a decrease in MAP because of the prevention of venous return in all groups. The slight increase of MAP in the SAH groups may reflect the preservation of Cushing's response.
Neither PC- nor VC-IRV during normocapnia led to any significant decrease in CPP in rabbits with raised ICP. Moreover, increases in PAW during inverse ratio ventilation appeared to affect ICP.
1. Burchiel KJ, Steege TD, Wyler AR. ICP changes in brain-injured patients requiring positive end-expiratory pressure ventilation. Neurosurgery
2. Hormann C, Mohsenipour I, Gottardis M, Benzer A. Response of cerebrospinal fluid pressure to continuous positive airway pressure in volunteers. Anesth Analg
3. Guerci AD, Shi AY, Levin H, Tsitlik J, Weisfeldt ML, Chandra N. Transmission of intrathoracic pressure to the intracranial space during cardiopulmonary resuscitation in dogs. Circ Res
4. Bouma GJ, Muizelaar JP, Bandoh K, Marmarou A. Blood pressure and ICP-volume dynamics in severe head injury: relationship with cerebral blood flow. J Neurosurg
5. Clarke JP. The effects of inverse ratio ventilation on ICP: a preliminary report. Intensive Care Med
6. Vollmer DG, Hongo K, Ogawa H, Tsukahara T, Kassell NF. A study of the effectiveness of the iron-chelating agent deferoxamine as vasospasm prophylaxis in a rabbit model of subarachnoid hemorrhage. Neurosurgery
7. Arthur AS, Fergus AH, Lanzino G, Mathys J, Kassell NF, Lee KS. Systemic administration of the iron chelator deferiprone attenuates subarachnoid hemorrhage-induced cerebral vasospasm in the rabbit. Neurosurgery
8. Donegan J. Physiology and metabolism of the brain and spinal cord. In: Newfield P, Cottreli JE, eds. Neuroanesthesia: Handbook of Clinical and Physiologic Essentials.
San Francisco, USA: Little Brown, 1991: 3-29.
9. Kuroda Y, Murakami M, Tsuruta J, Murakawa T, Sakabe T. Blood flow velocity of middle cerebral artery during prolonged anesthesia with halothane, isoflurane, and sevoflurane in humans. Anesthesiology
10. Todd MM, Drummond JC. A comparison of the cerebrovascular and metabolic effects of halothane and isoflurane in the cat. Anesthesiology
11. Chan K, Abraham E. Effects of inverse ratio ventilation on cardiorespiratory parameters in severe respiratory failure. Chest
12. Mang H, Kacmarek RM, Ritz R, Wilson RS, Kimball WP. Cardiorespiratory effects of volume- and pressure-controlled ventilation at various I/E ratios in an acute lung injury model. Am J Respir Crit Care Med
13. Ludwigs U, Klingstedt C, Baehrendtz S, Hedenstierna G. A comparison of pressure- and volume-controlled ventilation at different inspiratory to expiratory ratios. Acta Anaesthesiol Scand
14. Lessard MR, Guerot E, Lorino H, Lemaire F, Brochard L. Effects of pressure-controlled with different I:E ratios versus volume-controlled ventilation on respiratory mechanics, gas exchange, and hemodynamics in patients with adult respiratory distress syndrome. Anesthesiology
15. Ludwigs U, Philip A, Robertson B, Hedenstierna G. Pulmonary epithelial permeability. An animal study of inverse ratio ventilation and conventional mechanical ventilation. Chest
16. Ludwigs U, Klingstedt C, Baehrendtz S, Wegenius G, Hedenstierna G. A functional and morphologic analysis of pressure-controlled inverse ratio ventilation in oleic acid-induced lung injury. Chest
17. Feldman Z, Robertson CS, Contant CF, Gopinath SP, Grossman RG. Positive end expiratory pressure reduces intracranial compliance in the rabbit. J Neurosurg Anesthesiol
18. Abraham E, Yoshihara G. Cardiorespiratory effects of pressure controlled ventilation in severe respiratory failure. Chest
19. Abraham E, Yoshihara G. Cardiorespiratory effects of pressure controlled inverse ratio ventilation in severe respiratory failure Chest