The prone position has been shown in nonrandomized investigations to redistribute gravity-dependent consolidations initially localized in dorsal lung regions (1). This results in a reduction of intrapulmonary venous admixture of blood (2) and improved arterial oxygenation in most mechanically ventilated patients with acute lung injury (ALI) (3–5). Recently, improvement in arterial oxygenation has been attributed largely to a decrease in thoracoabdominal compliance caused by an increased rigidity of the chest wall in the prone position (4,6). This presumably augments distribution of ventilation to initially collapsed dorsal lung regions (4,6).
On the basis of these observations, the effect of the prone position on gas exchange should be further enhanced when the protrusion of the abdominal wall is restricted (6). However, simply turning a patient prone without any efforts to relieve the restriction of the thorax and abdomen has been claimed to reduce venous return and cardiac output, presumably by an increase in intraabdominal pressure (IAP) (3). In anesthetized patients, prone positioning reduces venous return (7,8) and cardiac output (9). Bradley and Bradley (10) demonstrated that an IAP increased to >15 mm Hg may considerably reduce renal perfusion and function in human volunteers (10). On the basis of these reports, further artificial restriction of abdominal expansion during prone positioning should accentuate these side effects, even in patients without abdominal hypertension. Although in critically ill patients at risk for multiple organ dysfunction, prone positioning without minimizing abdominal restriction may be advantageous for arterial oxygenation, its effects on IAP and cardiovascular and renal function have not been investigated.
We designed this study to detect any impairment in cardiovascular or renal function during prone positioning in critically ill patients with ALI when no attempt was made to restrict or enhance the movement of the abdominal wall.
After approval by the Bonn university ethics committee and informed written consent obtained from the family, we studied 16 mechanically ventilated patients with ALI. The criteria of the American-European Consensus Conference were used to define ALI (11). Patients were included within 24 h when they met the ALI criteria. Patients with unstable cardiovascular function requiring inotropic support, patients receiving diuretics, and those with renal transplants or renal replacement therapy were excluded, as were patients with cerebral injury, unstable spinal fractures, or those treated for generalized peritonitis with an open-abdomen technique. The Organ Failure Score (12), the Simplified Acute Physiology Score (13), and the duration of mechanical ventilation were recorded on inclusion in the study (Table 1).
Routine clinical management of the patients included the use of a central venous catheter and a thermistor-tipped fiberoptic arterial catheter (Pulsiocath PV2024-4F; Pulsion Medical Systems, Munich, Germany) advanced via the femoral artery into the aorta.
The heart rate was obtained from the electrocardiogram. Mean arterial pressure (MAP) and central venous pressure (CVP) were transduced (Combitrans®; Braun AG, Melsungen, Germany) and recorded. A horizontal plane through the midaxillary line was taken as a zero reference point for pressure measurements. The transpulmonary double-indicator dilution method was used to estimate cardiac output and intrathoracic blood volume, as described previously (14,15). Indocyanine green dye (ICG) (Becton Dickinson, Cockeysville, MD), 25 mg dissolved in 15 mL iced 5% glucose solution, was used as a double indicator and injected into the right atrium via the central venous line. Simultaneously, dilution curves for dye and temperature were recorded in the aorta with the thermistor-tipped fiberoptic arterial catheter. Cardiac output was estimated with the Stewart-Hamilton method (16), and the mean transit time of the first pass of the thermal and dye (mttICG) indicator was determined with a computer (COLD-Z-021; Pulsion Medical Systems, Munich, Germany). An average of three measurements were performed at random moments during the ventilatory cycle.
Arterial blood gases and pH were determined immediately after sampling in duplicate with standard blood gas electrodes (ABL 510; Radiometer, Copenhagen, Denmark). Hemoglobin concentration and oxygen saturation were measured from each sample with a spectrophotometer (OSM 3; Radiometer, Copenhagen, Denmark).
Urine was collected at 60-min intervals after air-washout of the urinary bladder by using a Foley catheter. Effective renal plasma flow (ERPF) and glomerular filtration rate (GFR) were determined by the steady-state clearance technique of paraaminohippuric acid (Merck, Inc., West Point, PA) and inulin (Laevosan GmbH, Linz, Austria), as described previously (17). Urine and plasma samples were analyzed for paraaminohippuric acid and inulin (DU-40®; Beckman Instruments, Inc., Brea, CA), sodium, creatinine (Synchron CX7; Beckman Instruments, Inc., Brea, CA), and osmolality (Osmometer; Knauer, Berlin, Germany).
IAP was estimated by transducing and recording the urinary bladder pressure (Combitrans®; Braun AG, Melsungen, Germany) during transient clamping of the Foley catheter, as described previously (18).
Intrathoracic blood volume index (ITBVI) was calculated as cardiac index (CI) × mttICG (14,15). Standard formulas were used to calculate CI, systemic vascular resistance index, and oxygen delivery index. ERPF and GFR were determined by using the standard clearance formula (17). Effective renal blood flow index (ERBFI) was calculated as ERPF · ([1 − hematocrit] · m2)− 1, GFR index (GFRI) as GFR · m− 2 (17), renal fraction of CI (RF) as ERBFI · CI− 1, renal vascular resistance index (RVRI) as MAP · ERBFI− 1 · 80,000, and filtration fraction (FF) as GFR · ERBF− 1 (17). Fractional sodium excretion, osmolar clearance, and free water clearance were calculated with standard formulas (17).
After inclusion in the study, the patients were placed in air-cushioned beds (MQ/VQ-TheraKair; KCI, Höchstadt, Germany) and sedated with infusions of sufentanil and midazolam to achieve a Ramsay Sedation Score of 5 (19). All patients were sufficiently resuscitated, as indicated by an ITBVI > 800 mL/m2. Fluid replacement and infusions of all drugs remained unchanged throughout the study.
Pressure-limited, time-cycled mechanical ventilation was provided with a standard ventilator (SV 300; Siemens, Erlangen, Germany). Appropriate ventilatory settings were determined by the physician responsible for the care of the patient, and they were maintained unchanged throughout the study.
Patients then were placed, in random order, supine and prone and stayed in each position for 180 min. A 60-min equilibration period followed the turning maneuver before measurements. Three sets of measurements were performed at 60-min intervals during each position and averaged.
In the prone position, the head was turned laterally and the arms were pronated and parallel to the body. No efforts were made to alleviate the positional restriction of the abdomen or the thorax.
Results are expressed as mean ± sd. Data were evaluated for normal distribution with Shapiro-Wilk’s W test. The data obtained at the three time points during the supine or prone position were analyzed with the one-way analysis of variance test. For comparison between the supine and prone position, the paired Student’s t-test was used. Correlations were calculated by using linear regression analysis. Differences were considered to be statistically significant if P < 0.05.
Patients were ventilated with a positive end-expiratory pressure of 11–21 cm H2O (15 ± 3 cm H2O), an upper-airway pressure limit of 25–34 cm H2O (29 ± 3 cm H2O), a ventilator rate of 10–27 breaths/min (19 ± 6 breaths/min), and a fraction of inspired oxygen (Fio2) of 0.35–0.8 (0.6 ± 0.2). The resulting minute ventilation remained essentially unchanged between the supine (7.2–14.2 L/min) (9.0 ± 1.9 L/min) and prone position (6.9–13.6 L/min) (8.6 ± 1.8 L/min). Similarly, mean airway pressure of 15–30 cm H2O (25 ± 6 cm H2O) did not change between interventions.
Changes in IAP are shown in Figure 1. The IAP was moderately increased in the supine position (12 ± 4 mm Hg) and increased to 14 ± 5 mm Hg during the prone position (P < 0.05). Regression analysis revealed a weak inverse correlation between changes in IAP and ERBFI (Fig. 2). Changes in IAP did not correlate with changes of any other variable recorded.
Cardiovascular and gas exchange variables are shown in Table 2. The prone position was associated with an increase in CI (P < 0.05) and MAP (P < 0.01), whereas heart rate, CVP, systemic vascular resistance index, and ITBVI remained unchanged. The prone position resulted in an increased Pao2/Fio2 and higher oxygen delivery index (P < 0.05). Hemoglobin, Paco2, and arterial pH did not change between interventions.
Variables reflecting renal perfusion and function are given in Table 3. RVRI was markedly increased and RF was reduced in the prone position. However, this was not associated with a reduction in ERBFI. Urine volume, GFRI, FF, fractional sodium excretion, osmolar clearance, and free water clearance did not change during prone positioning.
There was no statistically significant variation or trend in the tested variables during the supine or prone position.
This study was designed to evaluate the effects of the prone position on the cardiopulmonary and renal function in patients with ALI. To avoid the confounding influence of inotropic and diuretic support, we investigated patients with intact cardiovascular and renal function. When no effort was made to minimize restriction of the abdomen or the thorax, prone positioning improved oxygen delivery by effecting increases in both cardiac output and Pao2. A small increase in IAP during prone positioning coincided with increased renal vascular resistance but no adverse effects on renal perfusion or function.
The prone position is increasingly used to improve arterial oxygenation in patients with ALI when mechanical ventilation with high positive end-expiratory pressure and Fio2 results in minimally acceptable Pao2 (11). Arterial hypoxemia caused by venous admixture during ALI correlates directly with the quantity of nonaerated tissue observed by computer tomography in dependent lung regions adjacent to the diaphragm (20). Prone positioning is associated with a significant increase in the volume of aerated lung tissue observed by computer tomography, decrease in venous admixture, and improvement of arterial blood oxygenation in most patients with ALI (1–5). Pelosi et al. (4) recently observed that the improvement in Pao2 during prone positioning correlated with the decrease in thoracoabdominal compliance. In anesthetized, mechanically ventilated pigs, an increase in IAP during the prone position resulted in further improvement in Pao2 (6). These data support the contention that when turning patients with ALI prone, one should not attempt to minimize the restriction of the abdomen if a gain in arterial blood oxygenation is desired. However, turning anesthetized patients prone reduces venous return and cardiac output, presumably by an increase in IAP (7–9). Moreover, data from human volunteers suggest that an increase in IAP to >15 millimeters of mercury may reduce renal perfusion and function (10). Gattinoni et al. (21) observed that in the supine position, average IAP ranges from six millimeters of mercury in patients with intrapulmonary acute respiratory distress syndrome (ARDS) to 16 millimeters of mercury in patients with extrapulmonary ARDS. Therefore, prone positioning may promote renal dysfunction in critically ill patients despite arterial oxygenation improvements.
This controversy has been addressed by only a few studies in critically ill patients. Pelosi et al. (4) observed in patients with ALI essentially no change in IAP during prone positioning when free protrusion and motion of the abdomen was assured. Data from small, nonrandomized studies indicate that global cardiac output will not be altered by prone positioning regardless of whether abdominal wall movement is unrestricted (2,4) or restricted (5). In our patients, cardiac output increased, and all variables reflecting renal perfusion and function indicated no negative effects from prone positioning.
Apart from using air-cushioned beds, we made no effort to minimize restriction of the abdomen during prone positioning. Nevertheless, we observed only a small increase in IAP from 12 to 14 millimeters of mercury while the patients were prone. This finding was not expected, because it has been suggested that free expansion of the abdominal wall is essential during prone positioning to avoid side effects caused by a marked increase in IAP (2–4).
Prone positioning improved arterial blood oxygenation in this investigation. Prone positioning in nonrandomized trials decreased pulmonary blood flow to shunt units (2) and increased arterial blood oxygenation in patients with ALI or ARDS (3–5).
In our patients, prone positioning was accompanied with a consistent increase in CI at constant CVP and ITBVI, indicating unchanged cardiac preload. Our results contradict the reported decrease in venous return (7,8) and cardiac output (9) in anesthetized patients during prone positioning presumed to result from increased IAP. Yokoyama et al. (22) found in healthy patients undergoing anesthesia no change in cardiac output when no efforts were made to minimize restriction of the abdomen during prone positioning. Similar results have been documented in mechanically ventilated patients with ARDS (5). The difference between our study in critically ill patients and previous investigations performed in patients under general anesthesia may be explained by different preexisting conditions in circulatory volume status. Alterations of venous return and cardiac output depend on the intravascular fluid load, especially when IAP increases (23). In the preoperative and intraoperative phase, circulatory volume may change significantly and thereby may even be inadequate. Unfortunately, in previous studies performed perioperatively, it has not been documented whether patients were sufficiently resuscitated before positioning. In contrast, in our patients sufficient resuscitation of intravascular volume, as monitored by ITBVI, was required for study entry.
Not surprisingly, a higher CI and better arterial blood oxygenation in our patients were not associated with worsening of renal function, even when IAP was moderately increased. However, despite augmented systemic blood flow during prone positioning, ERBFI remained essentially unchanged, resulting in a lower RF. This phenomenon can be explained by renal autoregulation, which maintains renal perfusion constant over a wide arterial pressure range (24). This explanation is strongly supported by the unchanged GFRI and FF, which depend on glomerular perfusion pressure. Vasoconstriction induced by a myogenic mechanism of the afferent arterioles may have increased renal vascular resistance to maintain glomerular perfusion constant (24). It seems unlikely that the observed increase in RVRI during prone positioning was induced by reflex stimulation of the neurohormonal system, because CI and MAP increased while ITBVI remained unchanged. Another explanation for the increase in RVRI may be the release of local vasoactive substances associated with the compression of the renal parenchyma and vasculature secondary to increased IAP (25). The unchanged renal function, as reflected by urine output, glomerular filtration, excretion of sodium, and clearances of osmolar substances and free water in the presence of renal vasoconstriction during prone positioning, can be explained only by the maintenance of a sufficient renal blood flow caused by the increase in cardiac output. As a consequence of improved arterial blood oxygenation at a constant renal blood flow, oxygen supply to the renal tissue should have even increased during prone positioning. Prone positioning is not likely to affect renal function as long as cardiovascular function remains stable in patients with moderately increased IAP. However, we observed a weak inverse correlation between changes in IAP and ERBFI. Therefore, further studies are needed to test the hypothesis that prone positioning might worsen the situation in patients with preexisting abdominal hypertension.
The results of this study demonstrate that prone positioning in patients with ALI, although associated with a small increase in IAP, contributes to improved arterial oxygenation and systemic blood flow without affecting renal perfusion and function. Because resting patients on their thoraces and abdomens did not provide any disadvantage in cardiocirculatory and renal function, special support to allow free chest and abdominal movement seems unnecessary when mechanically ventilated, cardiovascularly stable patients with moderately increased IAP are turned prone to improve gas exchange.
The authors would like to thank the nursing staff for their valuable cooperation.
1. Gattinoni L, Pelosi P, Vitale G, et al. Body position changes redistribute lung computed tomographic density in patients with acute respiratory failure. Anesthesiology 1991; 74: 15–23.
2. Pappert D, Rossaint R, Slama K, et al. Influence of positioning on ventilation-perfusion relationships in severe adult respiratory distress syndrome. Chest 1994; 106: 1511–6.
3. Douglas WW, Rehder K, Froukje MB, et al. Improved oxygenation in patients with acute respiratory distress syndrome. Am Rev Respir Dis 1974; 115: 559–66.
4. Pelosi P, Tubiolo D, Mascheroni D, et al. Effects of the prone position on respiratory mechanics and gas exchange during acute lung injury. Am J Respir Crit Care Med 1998; 157: 387–93.
5. Blanch L, Mancebo J, Perez M, et al. Short-term effects of prone position in critically ill patients with acute respiratory distress syndrome. Intensive Care Med 1997; 23: 1033–9.
6. Mure M, Glenny RW, Domino KB, Hlastala MP. Pulmonary gas exchange in the prone position with abdominal distension. Am J Respir Crit Care Med 1998; 157: 1785–90.
7. Toyota S, Amaki Y. Hemodynamic evaluation of the prone position by transesophageal echocardiography. J Clin Anesth 1998; 10: 32–5.
8. Soliman DE, Maslow AD, Bokesch PM, et al. Transoesophageal echocardiography during scoliosis repair: comparison with CVP monitoring. Can J Anaesth 1998; 45: 925–32.
9. Backofen JE, Schauble JF. Hemodynamic changes with prone positioning during general anesthesia. Anesth Analg 1995; 64: 194.
10. Bradley SE, Bradley GP. The effect of increased intra-abdominal pressure on renal function in man. J Clin Invest 1947; 26: 1010–22.
11. Bernard GR, Artigas A, Brigham KL, et al. Report of the American-European consensus conference on ARDS: definitions, mechanisms, relevant outcomes and clinical trial coordination. Am J Respir Crit Care Med 1994; 149: 818–24.
12. Goris RJA, teBoekhorst TPA, Nuytinck JKS, Gimbere JKS. Multiple organ failure: generalized autodestructive inflammation? Arch Surg 1985; 120: 1109–15.
13. Le Gall JR, Loirat P, Alperovitch A, et al. A simplified acute physiology score for ICU patients. Crit Care Med 1984; 12: 975–7.
14. Hoeft A. Transpulmonary indicator dilution: an alternative approach for hemodynamic monitoring. In Vincent JL, ed. Yearbook of intensive care and emergency medicine. Berlin: Springer-Verlag, 1995: 593–605.
15. Godje O, Peyerl M, Seebauer T, et al. Reproducibility of double indicator dilution measurements of intrathoracic blood volume compartments, extravascular lung water, and liver function. Chest 1998; 113: 1070–7.
16. Stewart GN. The pulmonary circulation time, the quantity of blood in lungs and the output of the heart. Am J Physiol 1921; 58: 20–44.
17. Duarte CB, Preuss HG. Assessment of renal function: glomerular and tubular. Clin Lab Med 1993; 13: 33–52.
18. Iberti TJ, Lieber CE, Benjamin E. Determination of intra-abdominal pressure using a transurethral bladder catheter: clinical validation of the technique. Anesthesiology 1989; 70: 47–50.
19. Ramsay MAE, Savege TM, Simpson BRJ, Goodwin R. Controlled sedation with alphaxalone-alphadolone. BMJ 1974; 22: 656–9.
20. Gattinoni L, Mascheroni D, Torresin A, et al. Morphological response to positive end expiratory pressure in acute respiratory failure: computerized tomography study. Intensive Care Med 1986; 12: 137–42.
21. Gattinoni L, Pelosi P, Suter PM, et al. Acute respiratory distress syndrome caused by pulmonary and extrapulmonary disease: different syndromes? Am J Respir Crit Care Med 1998; 158: 3–11.
22. Yokoyama M, Ueda W, Hirakawa M, Yamamoto H. Haemodynamic effect of the prone position during anaesthesia. Acta Anaesthesiol Scand 1991; 35: 741–4.
23. Ridings PC, Bloomfield GL, Blocher CR, Sugerman HJ. Cardiopulmonary effects of raised intra-abdominal pressure before and after volume expansion. J Trauma 1995; 39: 1071–5.
24. Thurau K. Renal hemodynamics. Am J Med 1964; 36: 698.
© 2001 International Anesthesia Research Society
25. Hamilton BD, Chow GK, Inman SR, et al. Increased intra-abdominal pressure during pneumoperitoneum stimulates endothelin release in a canine model. J Endourol 1998; 12: 193–7.